Unveilling Pattern on Students’ Academic Success
Project on Association Rules Method
I Putu Agastya Harta Pratama
20 February 2025
Introduction
Understanding factors of student academic performance is something that could not be taken with a grain of salt, and understanding why students leave their programs has always been a complex issue. A Research by by Al Husaini and Ahmad Shukor (2023), has identified several key determinants, including low entry grades, family support, accommodation, gender, previous assessment grades, and e-learning activity, all of which significantly affect students’ academic success.
In order to discover this pattern, I found a dataset that tracks various student attributes — demographics, academic performance, and socioeconomic factors. Since this dataset is usually used for classification algorithm, thus it has a “Target” features within it.
Due to the underlying usage of this dataset, I structured my analysis into two phases. Phase 1 mostly focuses on exploring general patterns within student attributes, such as grades, attendance, and financial status, without immediately tying them to the target outcome. This helps me see which factors naturally group together and how students with similar characteristics behave. Then, in Phase 2, I shift the focus specifically to graduation and dropout outcomes, identifying which combinations of factors are most strongly linked to each.
Libraries Import
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(arules)## Loading required package: Matrix##
## Attaching package: 'arules'## The following object is masked from 'package:dplyr':
##
## recode## The following objects are masked from 'package:base':
##
## abbreviate, writelibrary(moments)
library(arulesViz)
library(arulesCBA)
library(VennDiagram)## Loading required package: grid## Loading required package: futile.loggerThe dataset that I will be using has 36 features and 4424
observations. It has various student attributes, including:
demographics, academic performance, and other socioeconomic factors.
Before preceeding, I will check whether the data is well-structure
enough to proceed to Association Rules algorithm.
Link to the
dataset: https://archive.ics.uci.edu/dataset/697/predict+students+dropout+and+academic+success
Data Import
data <- read.csv("predict_student_dropout.csv",header=TRUE,sep=";")
head(data, 7)## Marital.status Application.mode Application.order Course
## 1 1 17 5 171
## 2 1 15 1 9254
## 3 1 1 5 9070
## 4 1 17 2 9773
## 5 2 39 1 8014
## 6 2 39 1 9991
## 7 1 1 1 9500
## Daytime.evening.attendance. Previous.qualification
## 1 1 1
## 2 1 1
## 3 1 1
## 4 1 1
## 5 0 1
## 6 0 19
## 7 1 1
## Previous.qualification..grade. Nacionality Mother.s.qualification
## 1 122.0 1 19
## 2 160.0 1 1
## 3 122.0 1 37
## 4 122.0 1 38
## 5 100.0 1 37
## 6 133.1 1 37
## 7 142.0 1 19
## Father.s.qualification Mother.s.occupation Father.s.occupation
## 1 12 5 9
## 2 3 3 3
## 3 37 9 9
## 4 37 5 3
## 5 38 9 9
## 6 37 9 7
## 7 38 7 10
## Admission.grade Displaced Educational.special.needs Debtor
## 1 127.3 1 0 0
## 2 142.5 1 0 0
## 3 124.8 1 0 0
## 4 119.6 1 0 0
## 5 141.5 0 0 0
## 6 114.8 0 0 1
## 7 128.4 1 0 0
## Tuition.fees.up.to.date Gender Scholarship.holder Age.at.enrollment
## 1 1 1 0 20
## 2 0 1 0 19
## 3 0 1 0 19
## 4 1 0 0 20
## 5 1 0 0 45
## 6 1 1 0 50
## 7 1 0 1 18
## International Curricular.units.1st.sem..credited.
## 1 0 0
## 2 0 0
## 3 0 0
## 4 0 0
## 5 0 0
## 6 0 0
## 7 0 0
## Curricular.units.1st.sem..enrolled. Curricular.units.1st.sem..evaluations.
## 1 0 0
## 2 6 6
## 3 6 0
## 4 6 8
## 5 6 9
## 6 5 10
## 7 7 9
## Curricular.units.1st.sem..approved. Curricular.units.1st.sem..grade.
## 1 0 0.00000
## 2 6 14.00000
## 3 0 0.00000
## 4 6 13.42857
## 5 5 12.33333
## 6 5 11.85714
## 7 7 13.30000
## Curricular.units.1st.sem..without.evaluations.
## 1 0
## 2 0
## 3 0
## 4 0
## 5 0
## 6 0
## 7 0
## Curricular.units.2nd.sem..credited. Curricular.units.2nd.sem..enrolled.
## 1 0 0
## 2 0 6
## 3 0 6
## 4 0 6
## 5 0 6
## 6 0 5
## 7 0 8
## Curricular.units.2nd.sem..evaluations. Curricular.units.2nd.sem..approved.
## 1 0 0
## 2 6 6
## 3 0 0
## 4 10 5
## 5 6 6
## 6 17 5
## 7 8 8
## Curricular.units.2nd.sem..grade.
## 1 0.00000
## 2 13.66667
## 3 0.00000
## 4 12.40000
## 5 13.00000
## 6 11.50000
## 7 14.34500
## Curricular.units.2nd.sem..without.evaluations. Unemployment.rate
## 1 0 10.8
## 2 0 13.9
## 3 0 10.8
## 4 0 9.4
## 5 0 13.9
## 6 5 16.2
## 7 0 15.5
## Inflation.rate GDP Target
## 1 1.4 1.74 Dropout
## 2 -0.3 0.79 Graduate
## 3 1.4 1.74 Dropout
## 4 -0.8 -3.12 Graduate
## 5 -0.3 0.79 Graduate
## 6 0.3 -0.92 Graduate
## 7 2.8 -4.06 GraduatePreliminary Data Check
str(data)## 'data.frame': 4424 obs. of 37 variables:
## $ Marital.status : int 1 1 1 1 2 2 1 1 1 1 ...
## $ Application.mode : int 17 15 1 17 39 39 1 18 1 1 ...
## $ Application.order : int 5 1 5 2 1 1 1 4 3 1 ...
## $ Course : int 171 9254 9070 9773 8014 9991 9500 9254 9238 9238 ...
## $ Daytime.evening.attendance. : int 1 1 1 1 0 0 1 1 1 1 ...
## $ Previous.qualification : int 1 1 1 1 1 19 1 1 1 1 ...
## $ Previous.qualification..grade. : num 122 160 122 122 100 ...
## $ Nacionality : int 1 1 1 1 1 1 1 1 62 1 ...
## $ Mother.s.qualification : int 19 1 37 38 37 37 19 37 1 1 ...
## $ Father.s.qualification : int 12 3 37 37 38 37 38 37 1 19 ...
## $ Mother.s.occupation : int 5 3 9 5 9 9 7 9 9 4 ...
## $ Father.s.occupation : int 9 3 9 3 9 7 10 9 9 7 ...
## $ Admission.grade : num 127 142 125 120 142 ...
## $ Displaced : int 1 1 1 1 0 0 1 1 0 1 ...
## $ Educational.special.needs : int 0 0 0 0 0 0 0 0 0 0 ...
## $ Debtor : int 0 0 0 0 0 1 0 0 0 1 ...
## $ Tuition.fees.up.to.date : int 1 0 0 1 1 1 1 0 1 0 ...
## $ Gender : int 1 1 1 0 0 1 0 1 0 0 ...
## $ Scholarship.holder : int 0 0 0 0 0 0 1 0 1 0 ...
## $ Age.at.enrollment : int 20 19 19 20 45 50 18 22 21 18 ...
## $ International : int 0 0 0 0 0 0 0 0 1 0 ...
## $ Curricular.units.1st.sem..credited. : int 0 0 0 0 0 0 0 0 0 0 ...
## $ Curricular.units.1st.sem..enrolled. : int 0 6 6 6 6 5 7 5 6 6 ...
## $ Curricular.units.1st.sem..evaluations. : int 0 6 0 8 9 10 9 5 8 9 ...
## $ Curricular.units.1st.sem..approved. : int 0 6 0 6 5 5 7 0 6 5 ...
## $ Curricular.units.1st.sem..grade. : num 0 14 0 13.4 12.3 ...
## $ Curricular.units.1st.sem..without.evaluations.: int 0 0 0 0 0 0 0 0 0 0 ...
## $ Curricular.units.2nd.sem..credited. : int 0 0 0 0 0 0 0 0 0 0 ...
## $ Curricular.units.2nd.sem..enrolled. : int 0 6 6 6 6 5 8 5 6 6 ...
## $ Curricular.units.2nd.sem..evaluations. : int 0 6 0 10 6 17 8 5 7 14 ...
## $ Curricular.units.2nd.sem..approved. : int 0 6 0 5 6 5 8 0 6 2 ...
## $ Curricular.units.2nd.sem..grade. : num 0 13.7 0 12.4 13 ...
## $ Curricular.units.2nd.sem..without.evaluations.: int 0 0 0 0 0 5 0 0 0 0 ...
## $ Unemployment.rate : num 10.8 13.9 10.8 9.4 13.9 16.2 15.5 15.5 16.2 8.9 ...
## $ Inflation.rate : num 1.4 -0.3 1.4 -0.8 -0.3 0.3 2.8 2.8 0.3 1.4 ...
## $ GDP : num 1.74 0.79 1.74 -3.12 0.79 -0.92 -4.06 -4.06 -0.92 3.51 ...
## $ Target : chr "Dropout" "Graduate" "Dropout" "Graduate" ...summary(data)## Marital.status Application.mode Application.order Course
## Min. :1.000 Min. : 1.00 Min. :0.000 Min. : 33
## 1st Qu.:1.000 1st Qu.: 1.00 1st Qu.:1.000 1st Qu.:9085
## Median :1.000 Median :17.00 Median :1.000 Median :9238
## Mean :1.179 Mean :18.67 Mean :1.728 Mean :8857
## 3rd Qu.:1.000 3rd Qu.:39.00 3rd Qu.:2.000 3rd Qu.:9556
## Max. :6.000 Max. :57.00 Max. :9.000 Max. :9991
## Daytime.evening.attendance. Previous.qualification
## Min. :0.0000 Min. : 1.000
## 1st Qu.:1.0000 1st Qu.: 1.000
## Median :1.0000 Median : 1.000
## Mean :0.8908 Mean : 4.578
## 3rd Qu.:1.0000 3rd Qu.: 1.000
## Max. :1.0000 Max. :43.000
## Previous.qualification..grade. Nacionality Mother.s.qualification
## Min. : 95.0 Min. : 1.000 Min. : 1.00
## 1st Qu.:125.0 1st Qu.: 1.000 1st Qu.: 2.00
## Median :133.1 Median : 1.000 Median :19.00
## Mean :132.6 Mean : 1.873 Mean :19.56
## 3rd Qu.:140.0 3rd Qu.: 1.000 3rd Qu.:37.00
## Max. :190.0 Max. :109.000 Max. :44.00
## Father.s.qualification Mother.s.occupation Father.s.occupation Admission.grade
## Min. : 1.00 Min. : 0.00 Min. : 0.00 Min. : 95.0
## 1st Qu.: 3.00 1st Qu.: 4.00 1st Qu.: 4.00 1st Qu.:117.9
## Median :19.00 Median : 5.00 Median : 7.00 Median :126.1
## Mean :22.28 Mean : 10.96 Mean : 11.03 Mean :127.0
## 3rd Qu.:37.00 3rd Qu.: 9.00 3rd Qu.: 9.00 3rd Qu.:134.8
## Max. :44.00 Max. :194.00 Max. :195.00 Max. :190.0
## Displaced Educational.special.needs Debtor
## Min. :0.0000 Min. :0.00000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.00000 1st Qu.:0.0000
## Median :1.0000 Median :0.00000 Median :0.0000
## Mean :0.5484 Mean :0.01153 Mean :0.1137
## 3rd Qu.:1.0000 3rd Qu.:0.00000 3rd Qu.:0.0000
## Max. :1.0000 Max. :1.00000 Max. :1.0000
## Tuition.fees.up.to.date Gender Scholarship.holder Age.at.enrollment
## Min. :0.0000 Min. :0.0000 Min. :0.0000 Min. :17.00
## 1st Qu.:1.0000 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:19.00
## Median :1.0000 Median :0.0000 Median :0.0000 Median :20.00
## Mean :0.8807 Mean :0.3517 Mean :0.2484 Mean :23.27
## 3rd Qu.:1.0000 3rd Qu.:1.0000 3rd Qu.:0.0000 3rd Qu.:25.00
## Max. :1.0000 Max. :1.0000 Max. :1.0000 Max. :70.00
## International Curricular.units.1st.sem..credited.
## Min. :0.00000 Min. : 0.00
## 1st Qu.:0.00000 1st Qu.: 0.00
## Median :0.00000 Median : 0.00
## Mean :0.02486 Mean : 0.71
## 3rd Qu.:0.00000 3rd Qu.: 0.00
## Max. :1.00000 Max. :20.00
## Curricular.units.1st.sem..enrolled. Curricular.units.1st.sem..evaluations.
## Min. : 0.000 Min. : 0.000
## 1st Qu.: 5.000 1st Qu.: 6.000
## Median : 6.000 Median : 8.000
## Mean : 6.271 Mean : 8.299
## 3rd Qu.: 7.000 3rd Qu.:10.000
## Max. :26.000 Max. :45.000
## Curricular.units.1st.sem..approved. Curricular.units.1st.sem..grade.
## Min. : 0.000 Min. : 0.00
## 1st Qu.: 3.000 1st Qu.:11.00
## Median : 5.000 Median :12.29
## Mean : 4.707 Mean :10.64
## 3rd Qu.: 6.000 3rd Qu.:13.40
## Max. :26.000 Max. :18.88
## Curricular.units.1st.sem..without.evaluations.
## Min. : 0.0000
## 1st Qu.: 0.0000
## Median : 0.0000
## Mean : 0.1377
## 3rd Qu.: 0.0000
## Max. :12.0000
## Curricular.units.2nd.sem..credited. Curricular.units.2nd.sem..enrolled.
## Min. : 0.0000 Min. : 0.000
## 1st Qu.: 0.0000 1st Qu.: 5.000
## Median : 0.0000 Median : 6.000
## Mean : 0.5418 Mean : 6.232
## 3rd Qu.: 0.0000 3rd Qu.: 7.000
## Max. :19.0000 Max. :23.000
## Curricular.units.2nd.sem..evaluations. Curricular.units.2nd.sem..approved.
## Min. : 0.000 Min. : 0.000
## 1st Qu.: 6.000 1st Qu.: 2.000
## Median : 8.000 Median : 5.000
## Mean : 8.063 Mean : 4.436
## 3rd Qu.:10.000 3rd Qu.: 6.000
## Max. :33.000 Max. :20.000
## Curricular.units.2nd.sem..grade.
## Min. : 0.00
## 1st Qu.:10.75
## Median :12.20
## Mean :10.23
## 3rd Qu.:13.33
## Max. :18.57
## Curricular.units.2nd.sem..without.evaluations. Unemployment.rate
## Min. : 0.0000 Min. : 7.60
## 1st Qu.: 0.0000 1st Qu.: 9.40
## Median : 0.0000 Median :11.10
## Mean : 0.1503 Mean :11.57
## 3rd Qu.: 0.0000 3rd Qu.:13.90
## Max. :12.0000 Max. :16.20
## Inflation.rate GDP Target
## Min. :-0.800 Min. :-4.060000 Length:4424
## 1st Qu.: 0.300 1st Qu.:-1.700000 Class :character
## Median : 1.400 Median : 0.320000 Mode :character
## Mean : 1.228 Mean : 0.001969
## 3rd Qu.: 2.600 3rd Qu.: 1.790000
## Max. : 3.700 Max. : 3.510000Data Preprocessing
I have noticed that some features hasn’t been coded to factors
datatype. I will change those accordingly.
I am taking the label
provided from the UCI Machine Learning variables table.
marital.status_map <- c(
"1" = "Single",
"2" = "Married",
"3" = "Widower",
"4" = "Divorced",
"5" = "Facto Union",
"6" = "Legally Separated"
)
application.mode_map <- c(
"1" = "1st Phase - General Contingent",
"2" = "Ordinance No. 612/93",
"5" = "1st Phase - Special Contingent (Azores Island)",
"7" = "Holders of Other Higher Courses",
"10" = "Ordinance No. 854-B/99",
"15" = "International Student (Bachelor)",
"16" = "1st Phase - Special Contingent (Madeira Island)",
"17" = "2nd Phase - General Contingent",
"18" = "3rd Phase - General Contingent",
"26" = "Ordinance No. 533-A/99, Item b2) (Different Plan)",
"27" = "Ordinance No. 533-A/99, Item b3) (Other Institution)",
"39" = "Over 23 years old",
"42" = "Transfer",
"43" = "Change of Course",
"44" = "Technological Specialization Diploma Holders",
"51" = "Change of Institution/Course",
"53" = "Short Cycle Diploma Holders",
"57" = "Change of Institution/Course (International)"
)
course_map <- c(
"33" = "Biofuel Production Technologies",
"171" = "Animation and Multimedia Design",
"8014" = "Social Service (Evening Attendance)",
"9003" = "Agronomy",
"9070" = "Communication Design",
"9085" = "Veterinary Nursing",
"9119" = "Informatics Engineering",
"9130" = "Equinculture",
"9147" = "Management",
"9238" = "Social Service",
"9254" = "Tourism",
"9500" = "Nursing",
"9556" = "Oral Hygiene",
"9670" = "Advertising and Marketing Management",
"9773" = "Journalism and Communication",
"9853" = "Basic Education",
"9991" = "Management (Evening Attendance)"
)
daytime.evening.attendance_map <- c(
"1" = "Daytime", "0" = "Evening"
)
previous.qualification_map <- c(
"1" = "Secondary Education",
"2" = "Higher Education - Bachelor's Degree",
"3" = "Higher Education - Degree",
"4" = "Higher Education - Master's",
"5" = "Higher Education - Doctorate",
"6" = "Frequency of Higher Education",
"9" = "12th Year of Schooling - Not Completed",
"10" = "11th Year of Schooling - Not Completed",
"12" = "Other - 11th Year of Schooling",
"14" = "10th Year of Schooling",
"15" = "10th Year of Schooling - Not Completed",
"19" = "Basic Education 3rd Cycle (9th/10th/11th Year) or Equivalent",
"38" = "Basic Education 2nd Cycle (6th/7th/8th Year) or Equivalent",
"39" = "Technological Specialization Course",
"40" = "Higher Education - Degree (1st Cycle)",
"42" = "Professional Higher Technical Course",
"43" = "Higher Education - Master (2nd Cycle)"
)
nacionality_map <- c(
"1" = "Portuguese",
"2" = "German",
"6" = "Spanish",
"11" = "Italian",
"13" = "Dutch",
"14" = "English",
"17" = "Lithuanian",
"21" = "Angolan",
"22" = "Cape Verdean",
"24" = "Guinean",
"25" = "Mozambican",
"26" = "Santomean",
"32" = "Turkish",
"41" = "Brazilian",
"62" = "Romanian",
"100" = "Moldova (Republic of)",
"101" = "Mexican",
"103" = "Ukrainian",
"105" = "Russian",
"108" = "Cuban",
"109" = "Colombian"
)
mother.s.qualification_map <- c(
"1" = "Secondary Education - 12th Year of Schooling or Eq.",
"2" = "Higher Education - Bachelor's Degree",
"3" = "Higher Education - Degree",
"4" = "Higher Education - Master's",
"5" = "Higher Education - Doctorate",
"6" = "Frequency of Higher Education 9 - 12th Year of Schooling - Not Completed",
"9" = "12th Year of Schooling - Not Completed",
"10" = "11th Year of Schooling - Not Completed",
"11" = "7th Year (Old)",
"12" = "Other - 11th Year of Schooling",
"14" = "10th Year of Schooling",
"18" = "General Commerce Course",
"19" = "Basic Education 3rd Cycle (9th/10th/11th Year) or Equiv.",
"22" = "Technical-Professional Course",
"26" = "7th Year of Schooling",
"27" = "2nd Cycle of General High School",
"29" = "9th Year of Schooling - Not Completed",
"30" = "8th Year of Schooling",
"34" = "Unknown",
"35" = "Can't Read or Write",
"36" = "Can read without having a 4th year of schooling",
"37" = "Basic education 1st cycle (4th/5th year) or equiv.",
"38" = "Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.",
"39" = "Technological Specialization Course",
"40" = "Higher Education - Degree (1st Cycle)",
"41" = "Specialized higher studies course",
"42" = "Professional higher technical course",
"43" = "Higher Education - Master's (2nd Cycle)",
"44" = "Higher Education - Doctorate (3rd cycle)"
)
father.s.qualification_map <- c(
"1" = "Secondary Education - 12th Year of Schooling or Eq.",
"2" = "Higher Education - Bachelor's Degree",
"3" = "Higher Education - Degree",
"4" = "Higher Education - Master's",
"5" = "Higher Education - Doctorate",
"6" = "Frequency of Higher Education 9 - 12th Year of Schooling - Not Completed",
"9" = "12th Year of Schooling - Not Completed",
"10" = "11th Year of Schooling - Not Completed",
"11" = "7th Year (Old)",
"12" = "Other - 11th Year of Schooling",
"14" = "10th Year of Schooling",
"18" = "General Commerce Course",
"19" = "Basic Education 3rd Cycle (9th/10th/11th Year) or Equiv.",
"22" = "Technical-Professional Course",
"26" = "7th Year of Schooling",
"27" = "2nd Cycle of General High School",
"29" = "9th Year of Schooling - Not Completed",
"30" = "8th Year of Schooling",
"34" = "Unknown",
"35" = "Can't Read or Write",
"36" = "Can read without having a 4th year of schooling",
"37" = "Basic education 1st cycle (4th/5th year) or equiv.",
"38" = "Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.",
"39" = "Technological Specialization Course",
"40" = "Higher Education - Degree (1st Cycle)",
"41" = "Specialized higher studies course",
"42" = "Professional higher technical course",
"43" = "Higher Education - Master's (2nd Cycle)",
"44" = "Higher Education - Doctorate (3rd cycle)",
"13" = "2nd year complementary high school",
"20" = "Complementary High School Course",
"25" = "Complementary High School Course - not concluded",
"31" = "General Course of Administration and Commerce",
"33" = "Supplementary Accounting and Administration"
)
mother.s.occupation_map <- c(
"0" = "Student",
"1" = "Representatives of the Legislative Power and Executive Bodies, Directors, Directors and Executive Managers",
"2" = "Specialists in Intellectual and Scientific Activities",
"3" = "Intermediate Level Technicians and Professions",
"4" = "Administrative Staff",
"5" = "Personal Services, Security and Safety Workers and Sellers",
"6" = "Farmers and Skilled Workers in Agriculture, Fisheries and Forestry",
"7" = "Skilled Workers in Industry, Construction and Craftsmen",
"8" = "Installation and Machine Operators and Assembly Workers",
"9" = "Unskilled Workers",
"10" = "Armed Forces",
"90" = "Other Situation",
"99" = "(Blank)",
"122" = "Health professionals",
"123" = "teachers",
"125" = "Specialists in information and communication technologies (ICT)",
"131" = "Intermediate level science and engineering technicians and professions",
"132" = "Technicians and professionals, of intermediate level of health",
"134" = "Intermediate level technicians from legal, social, sports, cultural and similar services",
"141" = "Office workers, secretaries in general and data processing operators",
"143" = "Data, accounting, statistical, financial services and registry-related operators",
"144" = "Other administrative support staff",
"151" = "personal service workers",
"152" = "sellers",
"153" = "Personal care workers and the like",
"171" = "Skilled construction workers and the like, except electricians",
"173" = "Skilled workers in printing, precision instrument manufacturing, jewelers, artisans and the like",
"175" = "Workers in food processing, woodworking, clothing and other industries and crafts",
"191" = "cleaning workers",
"192" = "Unskilled workers in agriculture, animal production, fisheries and forestry",
"193" = "Unskilled workers in extractive industry, construction, manufacturing and transport",
"194" = "Meal preparation assistants"
)
father.s.occupation_map <- c(
"0" = "Student",
"1" = "Representatives of the Legislative Power and Executive Bodies, Directors, Directors and Executive Managers",
"2" = "Specialists in Intellectual and Scientific Activities",
"3" = "Intermediate Level Technicians and Professions",
"4" = "Administrative Staff",
"5" = "Personal Services, Security and Safety Workers and Sellers",
"6" = "Farmers and Skilled Workers in Agriculture, Fisheries and Forestry",
"7" = "Skilled Workers in Industry, Construction and Craftsmen",
"8" = "Installation and Machine Operators and Assembly Workers",
"9" = "Unskilled Workers",
"10" = "Armed Forces",
"90" = "Other Situation",
"99" = "(Blank)",
"122" = "Health professionals",
"123" = "teachers",
"125" = "Specialists in information and communication technologies (ICT)",
"131" = "Intermediate level science and engineering technicians and professions",
"132" = "Technicians and professionals, of intermediate level of health",
"134" = "Intermediate level technicians from legal, social, sports, cultural and similar services",
"141" = "Office workers, secretaries in general and data processing operators",
"143" = "Data, accounting, statistical, financial services and registry-related operators",
"144" = "Other administrative support staff",
"151" = "personal service workers",
"152" = "sellers",
"153" = "Personal care workers and the like",
"171" = "Skilled construction workers and the like, except electricians",
"173" = "Skilled workers in printing, precision instrument manufacturing, jewelers, artisans and the like",
"175" = "Workers in food processing, woodworking, clothing and other industries and crafts",
"191" = "cleaning workers",
"192" = "Unskilled workers in agriculture, animal production, fisheries and forestry",
"193" = "Unskilled workers in extractive industry, construction, manufacturing and transport",
"194" = "Meal preparation assistants",
"101" = "Armed Forces Officers",
"102" = "Armed Forces Sergeants",
"103" = "Other Armed Forces personnel",
"112" = "Directors of administrative and commercial services",
"114" = "Hotel, catering, trade and other services directors",
"121" = "Specialists in the physical sciences, mathematics, engineering and related techniques",
"124" = "Specialists in finance, accounting, administrative organization, public and commercial relations",
"135" = "Information and communication technology technicians",
"154" = "Protection and security services personnel",
"161" = "Market-oriented farmers and skilled agricultural and animal production workers",
"163" = "Farmers, livestock keepers, fishermen, hunters and gatherers, subsistence",
"172" = "Skilled workers in metallurgy, metalworking and similar",
"174" = "Skilled workers in electricity and electronics",
"181" = "Fixed plant and machine operators",
"182" = "assembly workers",
"183" = "Vehicle drivers and mobile equipment operators",
"195" = "Street vendors (except food) and street service providers"
)
binary_map <- c("1" = "Yes", "0" = "No")
gender_map <- c("1" = "Male", "0" = "Female")# Apply mappings to categorical columns
data$Marital.status <- as.character(marital.status_map[as.character(data$Marital.status)])
data$Application.mode <- as.character(application.mode_map[as.character(data$Application.mode)])
data$Course <- as.character(course_map[as.character(data$Course)])
data$Daytime.evening.attendance. <- as.character(daytime.evening.attendance_map[as.character(data$Daytime.evening.attendance.)])
data$Previous.qualification <- as.character(previous.qualification_map[as.character(data$Previous.qualification)])
data$Nacionality <- as.character(nacionality_map[as.character(data$Nacionality)])
data$Mother.s.qualification <- as.character(mother.s.qualification_map[as.character(data$Mother.s.qualification)])
data$Mother.s.occupation <- as.character(mother.s.occupation_map[as.character(data$Mother.s.occupation)])
data$Father.s.qualification <- as.character(father.s.qualification_map[as.character(data$Father.s.qualification)])
data$Father.s.occupation <- as.character(father.s.occupation_map[as.character(data$Father.s.occupation)])
data$Displaced <- as.character(binary_map[as.character(data$Displaced)])
data$Educational.special.needs <- as.character(binary_map[as.character(data$Educational.special.needs)])
data$Debtor <- as.character(binary_map[as.character(data$Debtor)])
data$Tuition.fees.up.to.date <- as.character(binary_map[as.character(data$Tuition.fees.up.to.date)])
data$Gender <- as.character(gender_map[as.character(data$Gender)])
data$Scholarship.holder <- as.character(binary_map[as.character(data$Scholarship.holder)])
data$International <- as.character(binary_map[as.character(data$International)])Checking Null Values
Checking null values in all variables within the dataset.
colSums(is.na(data))## Marital.status
## 0
## Application.mode
## 0
## Application.order
## 0
## Course
## 0
## Daytime.evening.attendance.
## 0
## Previous.qualification
## 0
## Previous.qualification..grade.
## 0
## Nacionality
## 0
## Mother.s.qualification
## 0
## Father.s.qualification
## 0
## Mother.s.occupation
## 0
## Father.s.occupation
## 0
## Admission.grade
## 0
## Displaced
## 0
## Educational.special.needs
## 0
## Debtor
## 0
## Tuition.fees.up.to.date
## 0
## Gender
## 0
## Scholarship.holder
## 0
## Age.at.enrollment
## 0
## International
## 0
## Curricular.units.1st.sem..credited.
## 0
## Curricular.units.1st.sem..enrolled.
## 0
## Curricular.units.1st.sem..evaluations.
## 0
## Curricular.units.1st.sem..approved.
## 0
## Curricular.units.1st.sem..grade.
## 0
## Curricular.units.1st.sem..without.evaluations.
## 0
## Curricular.units.2nd.sem..credited.
## 0
## Curricular.units.2nd.sem..enrolled.
## 0
## Curricular.units.2nd.sem..evaluations.
## 0
## Curricular.units.2nd.sem..approved.
## 0
## Curricular.units.2nd.sem..grade.
## 0
## Curricular.units.2nd.sem..without.evaluations.
## 0
## Unemployment.rate
## 0
## Inflation.rate
## 0
## GDP
## 0
## Target
## 0Checking Duplicated Data
check_duplicates <- duplicated(data)
data[check_duplicates, ]## [1] Marital.status
## [2] Application.mode
## [3] Application.order
## [4] Course
## [5] Daytime.evening.attendance.
## [6] Previous.qualification
## [7] Previous.qualification..grade.
## [8] Nacionality
## [9] Mother.s.qualification
## [10] Father.s.qualification
## [11] Mother.s.occupation
## [12] Father.s.occupation
## [13] Admission.grade
## [14] Displaced
## [15] Educational.special.needs
## [16] Debtor
## [17] Tuition.fees.up.to.date
## [18] Gender
## [19] Scholarship.holder
## [20] Age.at.enrollment
## [21] International
## [22] Curricular.units.1st.sem..credited.
## [23] Curricular.units.1st.sem..enrolled.
## [24] Curricular.units.1st.sem..evaluations.
## [25] Curricular.units.1st.sem..approved.
## [26] Curricular.units.1st.sem..grade.
## [27] Curricular.units.1st.sem..without.evaluations.
## [28] Curricular.units.2nd.sem..credited.
## [29] Curricular.units.2nd.sem..enrolled.
## [30] Curricular.units.2nd.sem..evaluations.
## [31] Curricular.units.2nd.sem..approved.
## [32] Curricular.units.2nd.sem..grade.
## [33] Curricular.units.2nd.sem..without.evaluations.
## [34] Unemployment.rate
## [35] Inflation.rate
## [36] GDP
## [37] Target
## <0 rows> (or 0-length row.names)Discreditisation
Check Distribution
Before binning, I am checking the distribution of each numerical variables mentioned above, and also their descriptive statistics.
continuous_vars <- c(
"Previous.qualification..grade.",
"Admission.grade",
"Curricular.units.1st.sem..grade.",
"Curricular.units.1st.sem..credited.",
"Curricular.units.1st.sem..enrolled.",
"Curricular.units.1st.sem..evaluations.",
"Curricular.units.1st.sem..approved.",
"Curricular.units.1st.sem..without.evaluations.",
"Curricular.units.2nd.sem..grade.",
"Curricular.units.2nd.sem..credited.",
"Curricular.units.2nd.sem..enrolled.",
"Curricular.units.2nd.sem..evaluations.",
"Curricular.units.2nd.sem..approved.",
"Curricular.units.2nd.sem..without.evaluations.",
"Unemployment.rate",
"Inflation.rate",
"GDP",
"Age.at.enrollment"
)
# Loop through variables and plot histograms
par(mfrow = c(3, 3)) # Arrange plots in a grid
for (var in continuous_vars) {
hist(data[[var]], main = paste("Histogram of", var), xlab = var, col = "lightblue", border = "black")
}
for (var in continuous_vars) {
cat("\nSummary of", var, ":\n")
print(summary(data[[var]]))
}##
## Summary of Previous.qualification..grade. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 95.0 125.0 133.1 132.6 140.0 190.0
##
## Summary of Admission.grade :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 95.0 117.9 126.1 127.0 134.8 190.0
##
## Summary of Curricular.units.1st.sem..grade. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 11.00 12.29 10.64 13.40 18.88
##
## Summary of Curricular.units.1st.sem..credited. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 0.00 0.00 0.71 0.00 20.00
##
## Summary of Curricular.units.1st.sem..enrolled. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 5.000 6.000 6.271 7.000 26.000
##
## Summary of Curricular.units.1st.sem..evaluations. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 6.000 8.000 8.299 10.000 45.000
##
## Summary of Curricular.units.1st.sem..approved. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 3.000 5.000 4.707 6.000 26.000
##
## Summary of Curricular.units.1st.sem..without.evaluations. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.1377 0.0000 12.0000
##
## Summary of Curricular.units.2nd.sem..grade. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.00 10.75 12.20 10.23 13.33 18.57
##
## Summary of Curricular.units.2nd.sem..credited. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.5418 0.0000 19.0000
##
## Summary of Curricular.units.2nd.sem..enrolled. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 5.000 6.000 6.232 7.000 23.000
##
## Summary of Curricular.units.2nd.sem..evaluations. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 6.000 8.000 8.063 10.000 33.000
##
## Summary of Curricular.units.2nd.sem..approved. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.000 2.000 5.000 4.436 6.000 20.000
##
## Summary of Curricular.units.2nd.sem..without.evaluations. :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0000 0.0000 0.1503 0.0000 12.0000
##
## Summary of Unemployment.rate :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 7.60 9.40 11.10 11.57 13.90 16.20
##
## Summary of Inflation.rate :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -0.800 0.300 1.400 1.228 2.600 3.700
##
## Summary of GDP :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -4.060000 -1.700000 0.320000 0.001969 1.790000 3.510000
##
## Summary of Age.at.enrollment :
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 17.00 19.00 20.00 23.27 25.00 70.00for (var in continuous_vars) {
cat("\nSkewness of", var, ":", skewness(data[[var]], na.rm = TRUE), "\n")
}##
## Skewness of Previous.qualification..grade. : 0.3127614
##
## Skewness of Admission.grade : 0.5304199
##
## Skewness of Curricular.units.1st.sem..grade. : -1.567614
##
## Skewness of Curricular.units.1st.sem..credited. : 4.167635
##
## Skewness of Curricular.units.1st.sem..enrolled. : 1.618492
##
## Skewness of Curricular.units.1st.sem..evaluations. : 0.9763055
##
## Skewness of Curricular.units.1st.sem..approved. : 0.7660026
##
## Skewness of Curricular.units.1st.sem..without.evaluations. : 8.20462
##
## Skewness of Curricular.units.2nd.sem..grade. : -1.313205
##
## Skewness of Curricular.units.2nd.sem..credited. : 4.633248
##
## Skewness of Curricular.units.2nd.sem..enrolled. : 0.7878463
##
## Skewness of Curricular.units.2nd.sem..evaluations. : 0.3363831
##
## Skewness of Curricular.units.2nd.sem..approved. : 0.3061755
##
## Skewness of Curricular.units.2nd.sem..without.evaluations. : 7.265236
##
## Skewness of Unemployment.rate : 0.2119791
##
## Skewness of Inflation.rate : 0.2522898
##
## Skewness of GDP : -0.3939346
##
## Skewness of Age.at.enrollment : 2.054292We can see that several features within the dataset have different
distributions and varying level of skewness, this does impact on how we
categorise them.
For roughly normal distribution variables, I will
implement equal width binning, which divides value into fixed intervals.
For skewed variables, I will implement equal frequency binning,
which is suited best when some values are more common than the other.
For variables with extreme outliers, I will implement quantile
based discreditisation.
Aside from the mentioned, I will implement
custom method of discreditisation.
Discreditisation Method
#Previous.qualification..grade. - EQUAL WIDTH
data$Previous.qualification..grade. <- cut(
data$Previous.qualification..grade.,
breaks = 3,
labels = c("Low", "Medium", "High"),
include.lowest = TRUE
)
#Admission.grade - EQUAL FREQ
data$Admission.grade <- discretize(
data$Admission.grade,
method = "frequency",
breaks = 3,
labels = c("Low", "Medium", "High")
)
# Age at Enrollment - EQUAL FREQ
data$Age.at.enrollment <- discretize(
data$Age.at.enrollment,
method = "frequency",
breaks = 3,
labels = c("Young", "Middle-Age", "Old")
)
# # Curricular Units 1st Credited - EQUAL FREQ
data$Curricular.units.1st.sem..credited. <- cut(
data$Curricular.units.1st.sem..credited.,
breaks = c(-Inf, 0, 5, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# Curricular Units 1st Enrolled - EQUAL FREQ
data$Curricular.units.1st.sem..enrolled. <- cut(
data$Curricular.units.1st.sem..enrolled.,
breaks = c(-Inf, 0, 6, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# Curricular Units 1st Evaluations - EQUAL FREQ
data$Curricular.units.1st.sem..evaluations. <- discretize(
data$Curricular.units.1st.sem..evaluations.,
method = "frequency",
breaks = 3,
labels = c("Low", "Medium", "High")
)
# Curricular Units 1st Approved - EQUAL FREQ
data$Curricular.units.1st.sem..approved. <- discretize(
data$Curricular.units.1st.sem..approved.,
method = "frequency",
breaks = 3,
labels = c("Little", "Moderate", "Many")
)
#Curricular.units.1st.sem..grade - QUANTILE BASED
data$Curricular.units.1st.sem..grade. <- cut(
data$Curricular.units.1st.sem..grade.,
breaks = quantile(data$Curricular.units.1st.sem..grade., probs = seq(0, 1, 0.33), na.rm = TRUE),
include.lowest = TRUE,
labels = c("Low", "Medium", "High")
)
# #Curricular.units.1st.sem..without eval - CUSTOM BINNING
data$Curricular.units.1st.sem..without.evaluations. <- cut(
data$Curricular.units.1st.sem..without.evaluations.,
breaks = c(-Inf, 0, 5, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# # Curricular Units 2nd Credited - EQUAL FREQ
data$Curricular.units.2nd.sem..credited. <- cut(
data$Curricular.units.2nd.sem..credited.,
breaks = c(-Inf, 0, 5, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# Curricular Units 2nd Enrolled - EQUAL FREQ
data$Curricular.units.2nd.sem..enrolled. <- cut(
data$Curricular.units.2nd.sem..enrolled.,
breaks = c(-Inf, 0, 6, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# Curricular Units 2nd Evaluations - EQUAL FREQ
data$Curricular.units.2nd.sem..evaluations. <- discretize(
data$Curricular.units.2nd.sem..evaluations.,
method = "frequency",
breaks = 3,
labels = c("Low", "Medium", "High")
)
# Curricular Units 2nd Approved - EQUAL FREQ
data$Curricular.units.2nd.sem..approved. <- discretize(
data$Curricular.units.2nd.sem..approved.,
method = "frequency",
breaks = 3,
labels = c("Little", "Moderate", "Many")
)
# Curricular.units.2nd.sem..grade. - QUANTILE BASED
data$Curricular.units.2nd.sem..grade. <- cut(
data$Curricular.units.2nd.sem..grade.,
breaks = quantile(data$Curricular.units.2nd.sem..grade., probs = seq(0, 1, 0.33), na.rm = TRUE),
include.lowest = TRUE,
labels = c("Low", "Medium", "High")
)
# # Curricular.units.2nd.sem..without eval - CUSTOM BIN
data$Curricular.units.2nd.sem..without.evaluations. <- cut(
data$Curricular.units.2nd.sem..without.evaluations.,
breaks = c(-Inf, 0, 5, Inf),
labels = c("None", "Few", "Many"),
include.lowest = TRUE
)
# Unemployment.rate - EQUAL WIDTH
data$Unemployment.rate <- cut(
data$Unemployment.rate,
breaks = 3,
labels = c("Low", "Medium", "High"),
include.lowest = TRUE
)
# Inflation Rate - EQUAL WIDTH
data$Inflation.rate <- cut(
data$Inflation.rate,
breaks = 3,
labels = c("Low", "Medium", "High"),
include.lowest = TRUE
)
# GDP - EQUAL WIDTH
data$GDP <- cut(
data$GDP,
breaks = 3,
labels = c("Low", "Medium", "High"),
include.lowest = TRUE
)# Check the discreditisation
for (i in continuous_vars){
print(paste("Table for:", i))
print(table(data[[i]]))
}## [1] "Table for: Previous.qualification..grade."
##
## Low Medium High
## 1306 2914 204
## [1] "Table for: Admission.grade"
##
## Low Medium High
## 1332 1596 1496
## [1] "Table for: Curricular.units.1st.sem..grade."
##
## Low Medium High
## 1466 1554 1363
## [1] "Table for: Curricular.units.1st.sem..credited."
##
## None Few Many
## 3847 336 241
## [1] "Table for: Curricular.units.1st.sem..enrolled."
##
## None Few Many
## 180 2967 1277
## [1] "Table for: Curricular.units.1st.sem..evaluations."
##
## Low Medium High
## 1206 1494 1724
## [1] "Table for: Curricular.units.1st.sem..approved."
##
## Little Moderate Many
## 1274 1156 1994
## [1] "Table for: Curricular.units.1st.sem..without.evaluations."
##
## None Few Many
## 4130 275 19
## [1] "Table for: Curricular.units.2nd.sem..grade."
##
## Low Medium High
## 1472 1586 1326
## [1] "Table for: Curricular.units.2nd.sem..credited."
##
## None Few Many
## 3894 394 136
## [1] "Table for: Curricular.units.2nd.sem..enrolled."
##
## None Few Many
## 180 2995 1249
## [1] "Table for: Curricular.units.2nd.sem..evaluations."
##
## Low Medium High
## 1322 1355 1747
## [1] "Table for: Curricular.units.2nd.sem..approved."
##
## Little Moderate Many
## 1467 1140 1817
## [1] "Table for: Curricular.units.2nd.sem..without.evaluations."
##
## None Few Many
## 4142 261 21
## [1] "Table for: Unemployment.rate"
##
## Low Medium High
## 1472 1803 1149
## [1] "Table for: Inflation.rate"
##
## Low Medium High
## 2144 893 1387
## [1] "Table for: GDP"
##
## Low Medium High
## 1349 1323 1752
## [1] "Table for: Age.at.enrollment"
##
## Young Middle-Age Old
## 1041 1832 1551Transaction Transformation
Factorisation of All Variables
Before proceeding to converting the dataset into transaction format, I need to
character_columns <- sapply(data, is.character)
data[, character_columns] <- lapply(data[, character_columns], as.factor)
numerical_columns <- sapply(data, is.numeric)
data[, numerical_columns] <- lapply(data[, numerical_columns], as.factor)## Warning in `[<-.data.frame`(`*tmp*`, , numerical_columns, value =
## list(structure(1L, levels = "5", class = "factor"), : provided 4424 variables
## to replace 1 variablesstr(data)## 'data.frame': 4424 obs. of 37 variables:
## $ Marital.status : Factor w/ 6 levels "Divorced","Facto Union",..: 5 5 5 5 4 4 5 5 5 5 ...
## $ Application.mode : Factor w/ 18 levels "1st Phase - General Contingent",..: 4 10 1 4 15 15 1 5 1 1 ...
## $ Application.order : Factor w/ 1 level "5": 1 1 1 1 1 1 1 1 1 1 ...
## $ Course : Factor w/ 17 levels "Advertising and Marketing Management",..: 3 16 6 9 15 11 12 16 14 14 ...
## $ Daytime.evening.attendance. : Factor w/ 2 levels "Daytime","Evening": 1 1 1 1 2 2 1 1 1 1 ...
## $ Previous.qualification : Factor w/ 17 levels "10th Year of Schooling",..: 16 16 16 16 16 6 16 16 16 16 ...
## $ Previous.qualification..grade. : Factor w/ 3 levels "Low","Medium",..: 1 3 1 1 1 2 2 1 2 2 ...
## $ Nacionality : Factor w/ 21 levels "Angolan","Brazilian",..: 15 15 15 15 15 15 15 15 16 15 ...
## $ Mother.s.qualification : Factor w/ 29 levels "10th Year of Schooling",..: 11 25 9 10 9 9 11 9 25 25 ...
## $ Father.s.qualification : Factor w/ 34 levels "10th Year of Schooling",..: 27 21 10 10 11 10 11 10 29 12 ...
## $ Mother.s.occupation : Factor w/ 32 levels "(Blank)","Administrative Staff",..: 18 10 29 18 29 29 22 29 29 2 ...
## $ Father.s.occupation : Factor w/ 46 levels "(Blank)","Administrative Staff",..: 42 17 42 17 42 33 3 42 42 33 ...
## $ Admission.grade : Factor w/ 3 levels "Low","Medium",..: 2 3 2 1 3 1 2 1 2 2 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 95 120 131 190
## ..- attr(*, "discretized:method")= chr "frequency"
## $ Displaced : Factor w/ 2 levels "No","Yes": 2 2 2 2 1 1 2 2 1 2 ...
## $ Educational.special.needs : Factor w/ 2 levels "No","Yes": 1 1 1 1 1 1 1 1 1 1 ...
## $ Debtor : Factor w/ 2 levels "No","Yes": 1 1 1 1 1 2 1 1 1 2 ...
## $ Tuition.fees.up.to.date : Factor w/ 2 levels "No","Yes": 2 1 1 2 2 2 2 1 2 1 ...
## $ Gender : Factor w/ 2 levels "Female","Male": 2 2 2 1 1 2 1 2 1 1 ...
## $ Scholarship.holder : Factor w/ 2 levels "No","Yes": 1 1 1 1 1 1 2 1 2 1 ...
## $ Age.at.enrollment : Factor w/ 3 levels "Young","Middle-Age",..: 2 2 2 2 3 3 1 3 2 1 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 17 19 22 70
## ..- attr(*, "discretized:method")= chr "frequency"
## $ International : Factor w/ 2 levels "No","Yes": 1 1 1 1 1 1 1 1 2 1 ...
## $ Curricular.units.1st.sem..credited. : Factor w/ 3 levels "None","Few","Many": 1 1 1 1 1 1 1 1 1 1 ...
## $ Curricular.units.1st.sem..enrolled. : Factor w/ 3 levels "None","Few","Many": 1 2 2 2 2 2 3 2 2 2 ...
## $ Curricular.units.1st.sem..evaluations. : Factor w/ 3 levels "Low","Medium",..: 1 1 1 2 3 3 3 1 2 3 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 0 7 9 45
## ..- attr(*, "discretized:method")= chr "frequency"
## $ Curricular.units.1st.sem..approved. : Factor w/ 3 levels "Little","Moderate",..: 1 3 1 3 2 2 3 1 3 2 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 0 4 6 26
## ..- attr(*, "discretized:method")= chr "frequency"
## $ Curricular.units.1st.sem..grade. : Factor w/ 3 levels "Low","Medium",..: 1 3 1 3 2 2 3 1 3 1 ...
## $ Curricular.units.1st.sem..without.evaluations.: Factor w/ 3 levels "None","Few","Many": 1 1 1 1 1 1 1 1 1 1 ...
## $ Curricular.units.2nd.sem..credited. : Factor w/ 3 levels "None","Few","Many": 1 1 1 1 1 1 1 1 1 1 ...
## $ Curricular.units.2nd.sem..enrolled. : Factor w/ 3 levels "None","Few","Many": 1 2 2 2 2 2 3 2 2 2 ...
## $ Curricular.units.2nd.sem..evaluations. : Factor w/ 3 levels "Low","Medium",..: 1 1 1 3 1 3 2 1 2 3 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 0 7 9 33
## ..- attr(*, "discretized:method")= chr "frequency"
## $ Curricular.units.2nd.sem..approved. : Factor w/ 3 levels "Little","Moderate",..: 1 3 1 2 3 2 3 1 3 1 ...
## ..- attr(*, "discretized:breaks")= num [1:4] 0 4 6 20
## ..- attr(*, "discretized:method")= chr "frequency"
## $ Curricular.units.2nd.sem..grade. : Factor w/ 3 levels "Low","Medium",..: 1 3 1 2 2 2 3 1 3 3 ...
## $ Curricular.units.2nd.sem..without.evaluations.: Factor w/ 3 levels "None","Few","Many": 1 1 1 1 1 2 1 1 1 1 ...
## $ Unemployment.rate : Factor w/ 3 levels "Low","Medium",..: 2 3 2 1 3 3 3 3 3 1 ...
## $ Inflation.rate : Factor w/ 3 levels "Low","Medium",..: 2 1 2 1 1 1 3 3 1 2 ...
## $ GDP : Factor w/ 3 levels "Low","Medium",..: 3 2 3 1 2 2 1 1 2 3 ...
## $ Target : Factor w/ 3 levels "Dropout","Enrolled",..: 1 3 1 3 3 3 3 1 3 1 ...Variables Selection
Since we will be working for two different phases, I will be including all of the features for the phase 2 and omit three variables in phase 1 (which are: application order, nationality, and target). The “target” variables dictates whether student graduate or not.
Variables for Phase 1
data_general <- data[, !colnames(data) %in% c("Application.order", "Nacionality", "Target")]Variables for Phase 2
data_with_target <- dataVerifying whether or not the required variables are included in the variables for both phases.
names(data_general)## [1] "Marital.status"
## [2] "Application.mode"
## [3] "Course"
## [4] "Daytime.evening.attendance."
## [5] "Previous.qualification"
## [6] "Previous.qualification..grade."
## [7] "Mother.s.qualification"
## [8] "Father.s.qualification"
## [9] "Mother.s.occupation"
## [10] "Father.s.occupation"
## [11] "Admission.grade"
## [12] "Displaced"
## [13] "Educational.special.needs"
## [14] "Debtor"
## [15] "Tuition.fees.up.to.date"
## [16] "Gender"
## [17] "Scholarship.holder"
## [18] "Age.at.enrollment"
## [19] "International"
## [20] "Curricular.units.1st.sem..credited."
## [21] "Curricular.units.1st.sem..enrolled."
## [22] "Curricular.units.1st.sem..evaluations."
## [23] "Curricular.units.1st.sem..approved."
## [24] "Curricular.units.1st.sem..grade."
## [25] "Curricular.units.1st.sem..without.evaluations."
## [26] "Curricular.units.2nd.sem..credited."
## [27] "Curricular.units.2nd.sem..enrolled."
## [28] "Curricular.units.2nd.sem..evaluations."
## [29] "Curricular.units.2nd.sem..approved."
## [30] "Curricular.units.2nd.sem..grade."
## [31] "Curricular.units.2nd.sem..without.evaluations."
## [32] "Unemployment.rate"
## [33] "Inflation.rate"
## [34] "GDP"names(data_with_target)## [1] "Marital.status"
## [2] "Application.mode"
## [3] "Application.order"
## [4] "Course"
## [5] "Daytime.evening.attendance."
## [6] "Previous.qualification"
## [7] "Previous.qualification..grade."
## [8] "Nacionality"
## [9] "Mother.s.qualification"
## [10] "Father.s.qualification"
## [11] "Mother.s.occupation"
## [12] "Father.s.occupation"
## [13] "Admission.grade"
## [14] "Displaced"
## [15] "Educational.special.needs"
## [16] "Debtor"
## [17] "Tuition.fees.up.to.date"
## [18] "Gender"
## [19] "Scholarship.holder"
## [20] "Age.at.enrollment"
## [21] "International"
## [22] "Curricular.units.1st.sem..credited."
## [23] "Curricular.units.1st.sem..enrolled."
## [24] "Curricular.units.1st.sem..evaluations."
## [25] "Curricular.units.1st.sem..approved."
## [26] "Curricular.units.1st.sem..grade."
## [27] "Curricular.units.1st.sem..without.evaluations."
## [28] "Curricular.units.2nd.sem..credited."
## [29] "Curricular.units.2nd.sem..enrolled."
## [30] "Curricular.units.2nd.sem..evaluations."
## [31] "Curricular.units.2nd.sem..approved."
## [32] "Curricular.units.2nd.sem..grade."
## [33] "Curricular.units.2nd.sem..without.evaluations."
## [34] "Unemployment.rate"
## [35] "Inflation.rate"
## [36] "GDP"
## [37] "Target"Transform Data into Transaction Format
# General dataset
transactions_general <- as(data_general, "transactions")
# Target-specific dataset
transactions_with_target <- as(data_with_target, "transactions")Checking first five transactions for the phase one dataset.
inspect(head(transactions_general,5))## items transactionID
## [1] {Marital.status=Single,
## Application.mode=2nd Phase - General Contingent,
## Course=Animation and Multimedia Design,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Mother.s.qualification=Basic Education 3rd Cycle (9th/10th/11th Year) or Equiv.,
## Father.s.qualification=Other - 11th Year of Schooling,
## Mother.s.occupation=Personal Services, Security and Safety Workers and Sellers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=Medium,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=None,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.1st.sem..grade.=Low,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=None,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Medium,
## Inflation.rate=Medium,
## GDP=High} 1
## [2] {Marital.status=Single,
## Application.mode=International Student (Bachelor),
## Course=Tourism,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=High,
## Mother.s.qualification=Secondary Education - 12th Year of Schooling or Eq.,
## Father.s.qualification=Higher Education - Degree,
## Mother.s.occupation=Intermediate Level Technicians and Professions,
## Father.s.occupation=Intermediate Level Technicians and Professions,
## Admission.grade=High,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=No,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=High,
## Inflation.rate=Low,
## GDP=Medium} 2
## [3] {Marital.status=Single,
## Application.mode=1st Phase - General Contingent,
## Course=Communication Design,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Mother.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Father.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Mother.s.occupation=Unskilled Workers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=Medium,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=No,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.1st.sem..grade.=Low,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Medium,
## Inflation.rate=Medium,
## GDP=High} 3
## [4] {Marital.status=Single,
## Application.mode=2nd Phase - General Contingent,
## Course=Journalism and Communication,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Mother.s.qualification=Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.,
## Father.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Mother.s.occupation=Personal Services, Security and Safety Workers and Sellers,
## Father.s.occupation=Intermediate Level Technicians and Professions,
## Admission.grade=Low,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=High,
## Curricular.units.2nd.sem..approved.=Moderate,
## Curricular.units.2nd.sem..grade.=Medium,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Low,
## Inflation.rate=Low,
## GDP=Low} 4
## [5] {Marital.status=Married,
## Application.mode=Over 23 years old,
## Course=Social Service (Evening Attendance),
## Daytime.evening.attendance.=Evening,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Mother.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Father.s.qualification=Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.,
## Mother.s.occupation=Unskilled Workers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=High,
## Displaced=No,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=High,
## Curricular.units.1st.sem..approved.=Moderate,
## Curricular.units.1st.sem..grade.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=Medium,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=High,
## Inflation.rate=Low,
## GDP=Medium} 5Checking first five transactions for the phase two dataset.
inspect(head(transactions_with_target,5))## items transactionID
## [1] {Marital.status=Single,
## Application.mode=2nd Phase - General Contingent,
## Application.order=5,
## Course=Animation and Multimedia Design,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Nacionality=Portuguese,
## Mother.s.qualification=Basic Education 3rd Cycle (9th/10th/11th Year) or Equiv.,
## Father.s.qualification=Other - 11th Year of Schooling,
## Mother.s.occupation=Personal Services, Security and Safety Workers and Sellers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=Medium,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=None,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.1st.sem..grade.=Low,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=None,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Medium,
## Inflation.rate=Medium,
## GDP=High,
## Target=Dropout} 1
## [2] {Marital.status=Single,
## Application.mode=International Student (Bachelor),
## Application.order=5,
## Course=Tourism,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=High,
## Nacionality=Portuguese,
## Mother.s.qualification=Secondary Education - 12th Year of Schooling or Eq.,
## Father.s.qualification=Higher Education - Degree,
## Mother.s.occupation=Intermediate Level Technicians and Professions,
## Father.s.occupation=Intermediate Level Technicians and Professions,
## Admission.grade=High,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=No,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=High,
## Inflation.rate=Low,
## GDP=Medium,
## Target=Graduate} 2
## [3] {Marital.status=Single,
## Application.mode=1st Phase - General Contingent,
## Application.order=5,
## Course=Communication Design,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Nacionality=Portuguese,
## Mother.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Father.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Mother.s.occupation=Unskilled Workers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=Medium,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=No,
## Gender=Male,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Low,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.1st.sem..grade.=Low,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Medium,
## Inflation.rate=Medium,
## GDP=High,
## Target=Dropout} 3
## [4] {Marital.status=Single,
## Application.mode=2nd Phase - General Contingent,
## Application.order=5,
## Course=Journalism and Communication,
## Daytime.evening.attendance.=Daytime,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Nacionality=Portuguese,
## Mother.s.qualification=Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.,
## Father.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Mother.s.occupation=Personal Services, Security and Safety Workers and Sellers,
## Father.s.occupation=Intermediate Level Technicians and Professions,
## Admission.grade=Low,
## Displaced=Yes,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=No,
## Age.at.enrollment=Middle-Age,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=High,
## Curricular.units.2nd.sem..approved.=Moderate,
## Curricular.units.2nd.sem..grade.=Medium,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=Low,
## Inflation.rate=Low,
## GDP=Low,
## Target=Graduate} 4
## [5] {Marital.status=Married,
## Application.mode=Over 23 years old,
## Application.order=5,
## Course=Social Service (Evening Attendance),
## Daytime.evening.attendance.=Evening,
## Previous.qualification=Secondary Education,
## Previous.qualification..grade.=Low,
## Nacionality=Portuguese,
## Mother.s.qualification=Basic education 1st cycle (4th/5th year) or equiv.,
## Father.s.qualification=Basic Education 2nd Cycle (6th/7th/8th Year) or Equiv.,
## Mother.s.occupation=Unskilled Workers,
## Father.s.occupation=Unskilled Workers,
## Admission.grade=High,
## Displaced=No,
## Educational.special.needs=No,
## Debtor=No,
## Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..enrolled.=Few,
## Curricular.units.1st.sem..evaluations.=High,
## Curricular.units.1st.sem..approved.=Moderate,
## Curricular.units.1st.sem..grade.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..enrolled.=Few,
## Curricular.units.2nd.sem..evaluations.=Low,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=Medium,
## Curricular.units.2nd.sem..without.evaluations.=None,
## Unemployment.rate=High,
## Inflation.rate=Low,
## GDP=Medium,
## Target=Graduate} 5itemFrequencyPlot(transactions_general, topN=20, type="absolute", main="Top 25 Most Frequent Items")
This bar chart above displays the top 25 most frequent items found in
the dataset, representing the most common attributes among students. The
x-axis lists these attributes, while the y-axis shows their absolute
frequency across all transactions. The most frequent attributes include
“Educational.special.needs=No”, “International=No”, and
“Daytime.evening.attendance.=Daytime”, indicating that the majority of
students do not have special educational needs, are not international
students, and attend daytime classes. Additionally, features such as
“Debtor=No” and “Tuition.fees.up.to.date=Yes” suggest that most students
have their tuition paid on time.
On the academic side,
“Curricular.units.2nd.sem..enrolled.=Few” and
“Curricular.units.1st.sem..enrolled.=Few” are relatively common, which
might indicate that many students enroll in only a few courses per
semester. This frequency plot helps in understanding the dominant
characteristics within the dataset and provides a foundation for
exploring patterns related to student performance and outcomes.
Apriori Algorithm
Apriori - First Phase
Since I did not set a specific consequent for the rules, a large
number of association rules were generated. To refine the results and
focus on more meaningful patterns, I need to adjust the support and
confidence parameters. By increasing the minimum support, I can filter
out rules that apply to only a small fraction of the dataset, ensuring
that the discovered relationships are more widely applicable.
Similarly, by raising the confidence threshold, I can prioritize rules
that have a stronger predictive power, reducing the likelihood of weak
or coincidental associations.
rules_general <- apriori(
transactions_general,
parameter = list(supp = 0.05, conf = 0.8, minlen = 2),
control = list(verbose = FALSE))inspect(sort(rules_general, by = "confidence", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Course=Communication Design} => {Daytime.evening.attendance.=Daytime} 0.05108499 1 0.05108499 1.122558 226
## [2] {Curricular.units.1st.sem..credited.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05447559 1 0.05447559 3.464370 241
## [3] {Curricular.units.1st.sem..credited.=Many} => {Educational.special.needs=No} 0.05447559 1 0.05447559 1.011662 241
## [4] {Course=Tourism} => {Daytime.evening.attendance.=Daytime} 0.05696203 1 0.05696203 1.122558 252
## [5] {Course=Tourism} => {Educational.special.needs=No} 0.05696203 1 0.05696203 1.011662 252
## [6] {Course=Management (Evening Attendance)} => {Daytime.evening.attendance.=Evening} 0.06057866 1 0.06057866 9.159420 268
## [7] {Course=Advertising and Marketing Management} => {Daytime.evening.attendance.=Daytime} 0.06057866 1 0.06057866 1.122558 268
## [8] {Course=Journalism and Communication} => {Daytime.evening.attendance.=Daytime} 0.07481917 1 0.07481917 1.122558 331
## [9] {Course=Veterinary Nursing} => {Daytime.evening.attendance.=Daytime} 0.07617541 1 0.07617541 1.122558 337
## [10] {Course=Social Service} => {Daytime.evening.attendance.=Daytime} 0.08024412 1 0.08024412 1.122558 355
## [11] {Course=Management} => {Daytime.evening.attendance.=Daytime} 0.08589512 1 0.08589512 1.122558 380
## [12] {Course=Nursing} => {Daytime.evening.attendance.=Daytime} 0.17314647 1 0.17314647 1.122558 766
## [13] {Application.mode=Over 23 years old} => {Age.at.enrollment=Old} 0.17744123 1 0.17744123 2.852353 785
## [14] {Inflation.rate=Medium} => {GDP=High} 0.20185353 1 0.20185353 2.525114 893
## [15] {Course=Communication Design,
## International=No} => {Daytime.evening.attendance.=Daytime} 0.05063291 1 0.05063291 1.122558 224
## [16] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.2nd.sem..enrolled.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05357143 1 0.05357143 3.464370 237
## [17] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.2nd.sem..enrolled.=Many} => {Curricular.units.1st.sem..evaluations.=High} 0.05357143 1 0.05357143 2.566125 237
## [18] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.2nd.sem..enrolled.=Many} => {Educational.special.needs=No} 0.05357143 1 0.05357143 1.011662 237
## [19] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.1st.sem..evaluations.=High} => {Curricular.units.1st.sem..enrolled.=Many} 0.05424955 1 0.05424955 3.464370 240
## [20] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.1st.sem..approved.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05402351 1 0.05402351 3.464370 239The first phase does not us new insight that much since it’s pretty much obvious. Many students in a vert particular courses, especially non-STEM, attend daytime classes - which is expected for them and pretty much confirms the program structure patterns. Moreover, there are other notable patterns highlight academic behaviors, such as students who earn many first-semester credits frequently enrolling in many courses or undergoing more evaluations,which I believe suggests that there’s a link between early academic performance and higher engagement.
A few rules seem less relevant to my analysis, such as the correlation between medium inflation rates and high GDP, which isn’t directly actionable for understanding student performance. Overall, these insights give me a broad overview of student behaviors, but they don’t yet help me identify the key predictors of dropout or graduation. That will be my main focus in Phase 2.
But before we continue to phase 2, I want to see if I could make any relevant rules more apparent by filtering out the weak and irrelevant ones. I’m doing this by introducting lift threshold within the parameter of the apriori rule. Since lift measures how much likely the rule consequent (RHS) occurs when rule antecedent (LHS). Thus by setting the lift to be above 1.2, we’re filtering out those rules who do not produce meaningful relationship.
rules_filtered <- subset(rules_general, lift > 1.2)inspect(sort(rules_filtered, by = "confidence", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Curricular.units.1st.sem..credited.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05447559 1 0.05447559 3.464370 241
## [2] {Course=Management (Evening Attendance)} => {Daytime.evening.attendance.=Evening} 0.06057866 1 0.06057866 9.159420 268
## [3] {Application.mode=Over 23 years old} => {Age.at.enrollment=Old} 0.17744123 1 0.17744123 2.852353 785
## [4] {Inflation.rate=Medium} => {GDP=High} 0.20185353 1 0.20185353 2.525114 893
## [5] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.2nd.sem..enrolled.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05357143 1 0.05357143 3.464370 237
## [6] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.2nd.sem..enrolled.=Many} => {Curricular.units.1st.sem..evaluations.=High} 0.05357143 1 0.05357143 2.566125 237
## [7] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.1st.sem..evaluations.=High} => {Curricular.units.1st.sem..enrolled.=Many} 0.05424955 1 0.05424955 3.464370 240
## [8] {Curricular.units.1st.sem..credited.=Many,
## Curricular.units.1st.sem..approved.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05402351 1 0.05402351 3.464370 239
## [9] {International=No,
## Curricular.units.1st.sem..credited.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05289331 1 0.05289331 3.464370 234
## [10] {Educational.special.needs=No,
## Curricular.units.1st.sem..credited.=Many} => {Curricular.units.1st.sem..enrolled.=Many} 0.05447559 1 0.05447559 3.464370 241
## [11] {International=No,
## Curricular.units.1st.sem..credited.=Many} => {Curricular.units.1st.sem..evaluations.=High} 0.05289331 1 0.05289331 2.566125 234
## [12] {Course=Tourism,
## Curricular.units.1st.sem..enrolled.=Few} => {Curricular.units.2nd.sem..enrolled.=Few} 0.05447559 1 0.05447559 1.477129 241
## [13] {Course=Tourism,
## Curricular.units.1st.sem..credited.=None} => {Curricular.units.1st.sem..enrolled.=Few} 0.05402351 1 0.05402351 1.491068 239
## [14] {Course=Tourism,
## Curricular.units.2nd.sem..credited.=None} => {Curricular.units.1st.sem..enrolled.=Few} 0.05424955 1 0.05424955 1.491068 240
## [15] {Course=Tourism,
## Curricular.units.1st.sem..credited.=None} => {Curricular.units.2nd.sem..enrolled.=Few} 0.05402351 1 0.05402351 1.477129 239
## [16] {Course=Tourism,
## Curricular.units.2nd.sem..credited.=None} => {Curricular.units.2nd.sem..enrolled.=Few} 0.05424955 1 0.05424955 1.477129 240
## [17] {Course=Management (Evening Attendance),
## Age.at.enrollment=Old} => {Daytime.evening.attendance.=Evening} 0.05877034 1 0.05877034 9.159420 260
## [18] {Course=Management (Evening Attendance),
## Curricular.units.1st.sem..enrolled.=Few} => {Daytime.evening.attendance.=Evening} 0.05018083 1 0.05018083 9.159420 222
## [19] {Course=Management (Evening Attendance),
## Curricular.units.2nd.sem..enrolled.=Few} => {Daytime.evening.attendance.=Evening} 0.05040687 1 0.05040687 9.159420 223
## [20] {Course=Management (Evening Attendance),
## Scholarship.holder=No} => {Daytime.evening.attendance.=Evening} 0.05583183 1 0.05583183 9.159420 247Even after doing the filtering, I can see that some of the trivial
relationship are still present. For instance, if a student is enrolled
in an Evening Attendance program, of course, they will attend in the
evening. But rules that highlight academic behaviours, such as
curricular units credited, enrolled, and evaluated for the first and
second semester. From here we can say that, the rulee suggest that
high-performing students may take on heavier workloads.
plot(rules_filtered[1:10], method = "graph", engine = "htmlwidget")rules_general_top <- head(sort(rules_filtered, by="confidence"), 50)
plot(rules_general_top, method = "grouped", control = list(k = 10))
Apriori - Second Phase: Target of Dropout
Now we move on to generating rules for the the second phase.
rules_dropout <- apriori(
transactions_with_target,
parameter = list(supp = 0.05, conf = 0.8, minlen = 2),
appearance = list(rhs = "Target=Dropout", default = "lhs"),
control = list(verbose = FALSE))length(rules_dropout)## [1] 73966inspect(sort(rules_dropout, by = "confidence", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05244123 0.9830508 0.05334539 3.060533 232
## [2] {Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05244123 0.9830508 0.05334539 3.060533 232
## [3] {Application.order=5,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05244123 0.9830508 0.05334539 3.060533 232
## [4] {Application.order=5,
## Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05244123 0.9830508 0.05334539 3.060533 232
## [5] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [6] {Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [7] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [8] {Nacionality=Portuguese,
## Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [9] {Application.order=5,
## Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [10] {Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [11] {Application.order=5,
## Tuition.fees.up.to.date=No,
## Scholarship.holder=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.05108499 0.9826087 0.05198915 3.059156 226
## [12] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [13] {Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [14] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [15] {Nacionality=Portuguese,
## Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [16] {Application.order=5,
## Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [17] {Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [18] {Application.order=5,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [19] {Nacionality=Portuguese,
## Educational.special.needs=No,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223
## [20] {Application.order=5,
## Nacionality=Portuguese,
## Tuition.fees.up.to.date=No,
## Age.at.enrollment=Old,
## International=No,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.05040687 0.9823789 0.05131103 3.058441 223inspect(sort(rules_dropout, by = "support", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1844485 0.8007851 0.2303345 2.493085 816
## [2] {Application.order=5,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1844485 0.8007851 0.2303345 2.493085 816
## [3] {Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1828662 0.8025794 0.2278481 2.498671 809
## [4] {Application.order=5,
## Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1828662 0.8025794 0.2278481 2.498671 809
## [5] {Scholarship.holder=No,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.1808318 0.8000000 0.2260398 2.490640 800
## [6] {Application.order=5,
## Scholarship.holder=No,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.1808318 0.8000000 0.2260398 2.490640 800
## [7] {Nacionality=Portuguese,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [8] {Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [9] {Nacionality=Portuguese,
## Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [10] {Application.order=5,
## Nacionality=Portuguese,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [11] {Application.order=5,
## Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [12] {Application.order=5,
## Nacionality=Portuguese,
## Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1803797 0.8020101 0.2249096 2.496898 798
## [13] {Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1794756 0.8004032 0.2242315 2.491896 794
## [14] {Application.order=5,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1794756 0.8004032 0.2242315 2.491896 794
## [15] {Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.1790235 0.8016194 0.2233273 2.495682 792
## [16] {Application.order=5,
## Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Little,
## Curricular.units.2nd.sem..grade.=Low} => {Target=Dropout} 0.1790235 0.8016194 0.2233273 2.495682 792
## [17] {Nacionality=Portuguese,
## Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1787975 0.8038618 0.2224231 2.502663 791
## [18] {Educational.special.needs=No,
## Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1787975 0.8038618 0.2224231 2.502663 791
## [19] {Nacionality=Portuguese,
## Educational.special.needs=No,
## Scholarship.holder=No,
## International=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1787975 0.8038618 0.2224231 2.502663 791
## [20] {Application.order=5,
## Nacionality=Portuguese,
## Educational.special.needs=No,
## Scholarship.holder=No,
## Curricular.units.1st.sem..approved.=Little,
## Curricular.units.2nd.sem..approved.=Little} => {Target=Dropout} 0.1787975 0.8038618 0.2224231 2.502663 791The association rules predicting dropout show me that financial instability, academic struggles, and age play a significant role in whether a student is at risk. The most frequent pattern I found indicates that students who aren’t up to date on tuition fees, don’t receive scholarships, are older, and have performed poorly in their second semester are highly likely to drop out. The high confidence values (≥98%) tell me that when these conditions co-occur, dropout is almost certain. I also noticed that nationality (Portuguese students) and having no special educational needs appear in several rules, though I’d need to explore further to fully understand their impact. Since the lift values (~3.06) suggest that these factors increase dropout likelihood well beyond random chance, I can see that financial and academic struggles are clear risk indicators.
I also found that students who applied as their fifth choice, those who weren’t international, and those with low second-semester grades have a strong link to dropping out. This tells me that choosing a program as a lower preference and struggling academically later on are important signals of dropout risk. The insights I’ve gained suggest that financial aid, academic support, and better early engagement strategies could help students at risk.
Moving on, I want to check on how these factors compare to those leading to graduation.
plot(rules_dropout[1:10], method = "graph", engine="htmlwidget")rules_dropout_top <- head(sort(rules_dropout, by="confidence"), 50)
plot(rules_dropout_top, method = "grouped", control = list(k = 10))
From the plot above, it reveals that students who lack scholarships,
have low curricular unit grades, and earn few or no credits in both
semesters are highly associated with dropout. Additionally, students who
are Portuguese, not international, and have little academic engagement
(low enrollment and few approvals) also appear frequently in
dropout-related rules. This suggests that academic performance and
financial support play a crucial role in student retention, with those
struggling academically and lacking financial aid being the most at risk
of leaving their studies.
Apriori - Second Phase: Target of Graduate
We are now applying apriori algorithm and setting “Graduate” as the rule consequent.
rules_graduate <- apriori(
transactions_with_target,
parameter = list(supp = 0.05, conf = 0.8, minlen = 2),
appearance = list(rhs = "Target=Graduate", default = "lhs"),
control = list(verbose = FALSE)
)length(rules_graduate)## [1] 657750inspect(sort(rules_graduate, by = "confidence", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Previous.qualification..grade.=Medium,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06193490 0.9681979 0.06396926 1.939026 274
## [2] {Previous.qualification..grade.=Medium,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06125678 0.9678571 0.06329114 1.938343 271
## [3] {Previous.qualification..grade.=Medium,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.06125678 0.9678571 0.06329114 1.938343 271
## [4] {Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.06690778 0.9673203 0.06916817 1.937268 296
## [5] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07346293 0.9672619 0.07594937 1.937151 325
## [6] {Previous.qualification..grade.=Medium,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High} => {Target=Graduate} 0.05266727 0.9668050 0.05447559 1.936236 233
## [7] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.07820976 0.9664804 0.08092224 1.935586 346
## [8] {Previous.qualification..grade.=Medium,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.05176311 0.9662447 0.05357143 1.935114 229
## [9] {Previous.qualification..grade.=Medium,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..grade.=High,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High} => {Target=Graduate} 0.05153707 0.9661017 0.05334539 1.934827 228
## [10] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07707957 0.9660057 0.07979204 1.934635 341
## [11] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## International=No,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07707957 0.9660057 0.07979204 1.934635 341
## [12] {Daytime.evening.attendance.=Daytime,
## Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07662749 0.9658120 0.07933996 1.934247 339
## [13] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..credited.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07617541 0.9656160 0.07888788 1.933855 337
## [14] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..credited.=None,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.07572333 0.9654179 0.07843580 1.933458 335
## [15] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..evaluations.=Medium,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.05018083 0.9652174 0.05198915 1.933056 222
## [16] {Tuition.fees.up.to.date=Yes,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.07414105 0.9647059 0.07685353 1.932032 328
## [17] {Tuition.fees.up.to.date=Yes,
## Gender=Female,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..evaluations.=Medium,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06781193 0.9646302 0.07029837 1.931881 300
## [18] {Previous.qualification..grade.=Medium,
## Scholarship.holder=Yes,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06148282 0.9645390 0.06374322 1.931698 272
## [19] {Application.order=5,
## Previous.qualification..grade.=Medium,
## Scholarship.holder=Yes,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06148282 0.9645390 0.06374322 1.931698 272
## [20] {Previous.qualification..grade.=Medium,
## Scholarship.holder=Yes,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..approved.=Many,
## Curricular.units.2nd.sem..grade.=High,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.06125678 0.9644128 0.06351718 1.931445 271inspect(sort(rules_graduate, by = "support", decreasing = TRUE)[1:20])## lhs rhs support confidence coverage lift count
## [1] {Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3508137 0.8112912 0.4324141 1.624786 1552
## [2] {Application.order=5,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3508137 0.8112912 0.4324141 1.624786 1552
## [3] {Educational.special.needs=No,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3471971 0.8114105 0.4278933 1.625025 1536
## [4] {Application.order=5,
## Educational.special.needs=No,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3471971 0.8114105 0.4278933 1.625025 1536
## [5] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [6] {Tuition.fees.up.to.date=Yes,
## International=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [7] {Nacionality=Portuguese,
## Tuition.fees.up.to.date=Yes,
## International=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [8] {Application.order=5,
## Nacionality=Portuguese,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [9] {Application.order=5,
## Tuition.fees.up.to.date=Yes,
## International=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [10] {Application.order=5,
## Nacionality=Portuguese,
## Tuition.fees.up.to.date=Yes,
## International=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3431284 0.8117647 0.4226944 1.625734 1518
## [11] {Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.3415461 0.8167568 0.4181736 1.635732 1511
## [12] {Application.order=5,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.3415461 0.8167568 0.4181736 1.635732 1511
## [13] {Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.3413201 0.8023379 0.4254069 1.606855 1510
## [14] {Application.order=5,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..without.evaluations.=None,
## Curricular.units.2nd.sem..without.evaluations.=None} => {Target=Graduate} 0.3413201 0.8023379 0.4254069 1.606855 1510
## [15] {Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..without.evaluations.=None} => {Target=Graduate} 0.3401899 0.8179348 0.4159132 1.638091 1505
## [16] {Application.order=5,
## Tuition.fees.up.to.date=Yes,
## Curricular.units.1st.sem..approved.=Many,
## Curricular.units.1st.sem..without.evaluations.=None} => {Target=Graduate} 0.3401899 0.8179348 0.4159132 1.638091 1505
## [17] {Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.3397378 0.8271877 0.4107143 1.656622 1503
## [18] {Application.order=5,
## Curricular.units.2nd.sem..approved.=Many} => {Target=Graduate} 0.3397378 0.8271877 0.4107143 1.656622 1503
## [19] {Debtor=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3397378 0.8111171 0.4188517 1.624437 1503
## [20] {Application.order=5,
## Debtor=No,
## Curricular.units.1st.sem..approved.=Many} => {Target=Graduate} 0.3397378 0.8111171 0.4188517 1.624437 1503The association rules predicting graduation show me that consistent academic performance, financial stability, and prior qualification grades play a key role in student success. The most frequent pattern I found suggests that students who paid their tuition on time, received scholarships, had medium-level previous qualification grades, and performed well in both semesters are highly likely to graduate. The high confidence values (≥96%) tell me that when these conditions are met, graduation is almost guaranteed. Additionally, students who had no missing evaluations, medium first-semester evaluations, and many approved second-semester courses also show strong associations with graduation. Since the lift values (~1.93) suggest that these factors significantly improve graduation likelihood, I can see that academic engagement and financial support are strong predictors of success.
I also noticed that female students who received scholarships, performed consistently, and had no missing coursework evaluations were strongly linked to graduation. This tells us that scholarship programs and structured academic support may be key contributors to student success. Comparing this to the dropout rules, it’s clear that students with financial security and stable academic performance throughout their coursework have a much higher chance of graduating.
The insights I have gained, could suggest that universities should focus on targeted financial aid programs and academic tracking to support students at risk of dropout.
plot(rules_graduate[1:10], method = "graph", engine="htmlwidget")rules_graduate_top <- head(sort(rules_graduate, by="confidence"), 50)
plot(rules_graduate_top, method = "grouped", control = list(k = 10))
From the plot above, I can see that factors such as tuition fees
being up-to-date, previous qualification grades being medium, high
curricular unit grades, and scholarship holding play a significant role
in graduation outcomes.
Additionally, demographic aspects like being
Portuguese, female, and having no educational special needs also appear
frequently in rules leading to graduation. This confirms that a mix of
financial stability, academic performance, and background
characteristics are key indicators of student success.
Conclusion
With all things considered, the project gave insightful information about the variables affecting student outcomes and suggested possible areas for institutions to intervene in order to increase retention rates. Institutions can improve student success by taking proactive measures like financial aid programs, academic support systems, and early dropout risk detection if they have a better grasp of these trends. Another research also shows that institutional support and engagement are essential in reducing university dropout rates. A systematic review highlighted that vocational guidance, academic support, and strong institutional backing are vital for improving student retention and success.
References
Al Husaini, A., & Ahmad Shukor, S. (2023). Factors affecting students’ academic performance: A review. ResearchGate. Retrieved February 21, 2025, from https://www.researchgate.net/publication/367360842_Factors_Affecting_Students%27_Academic_Performance_A_review
Scaler. (n.d.). Binning in data mining. Scaler Topics. Retrieved February 21, 2025, from https://www.scaler.com/topics/binning-in-data-mining
Quincho Apumayta, R., Carrillo Cayllahua, J., Ccencho Pari, A., Inga Choque, V., Cárdenas Valverde, J. C., & Huamán Ataypoma, D. (2024). University dropout: A systematic review of the main determinant factors (2020–2024). F1000Research, 13, 942. https://doi.org/10.12688/f1000research.154263.2