Association Homework
Mustafa Kaan Koc
2025-02-05
library(dplyr)
library(arules)
library(arulesViz)
#load the dataset
data<-read.csv("ESS11.csv")
#examine dataset
str(data)## 'data.frame': 40156 obs. of 640 variables:
## $ name : chr "ESS11e02" "ESS11e02" "ESS11e02" "ESS11e02" ...
## $ essround : int 11 11 11 11 11 11 11 11 11 11 ...
## $ edition : num 2 2 2 2 2 2 2 2 2 2 ...
## $ proddate : chr "20.11.2024" "20.11.2024" "20.11.2024" "20.11.2024" ...
## $ idno : int 50014 50030 50057 50106 50145 50158 50211 50212 50213 50235 ...
## $ cntry : chr "AT" "AT" "AT" "AT" ...
## $ dweight : num 1.185 0.61 1.392 0.556 0.723 ...
## $ pspwght : num 0.393 0.325 4 0.176 1.061 ...
## $ pweight : num 0.331 0.331 0.331 0.331 0.331 ...
## $ anweight : num 0.13 0.1076 1.3237 0.0583 0.3511 ...
## $ nwspol : int 90 90 30 15 60 120 45 120 20 120 ...
## $ netusoft : int 5 5 5 1 5 5 1 5 5 1 ...
## $ netustm : int 180 570 30 6666 120 120 6666 240 180 6666 ...
## $ ppltrst : int 5 10 6 6 6 8 7 8 7 3 ...
## $ pplfair : int 5 0 9 6 3 8 7 8 8 4 ...
## $ pplhlp : int 5 1 8 6 8 4 8 8 7 3 ...
## $ polintr : int 1 2 2 3 2 2 2 1 3 2 ...
## $ psppsgva : int 4 3 4 2 3 2 2 4 2 3 ...
## $ actrolga : int 5 2 4 2 1 3 2 4 2 2 ...
## $ psppipla : int 4 3 4 2 4 2 2 4 2 3 ...
## $ cptppola : int 5 2 3 3 3 3 2 5 2 3 ...
## $ trstprl : int 6 6 7 5 6 3 6 9 6 8 ...
## $ trstlgl : int 9 6 5 6 8 5 5 9 7 8 ...
## $ trstplc : int 10 4 8 9 8 7 8 9 8 10 ...
## $ trstplt : int 5 1 4 3 5 5 4 7 3 8 ...
## $ trstprt : int 5 0 4 3 5 5 5 6 2 8 ...
## $ trstep : int 5 5 7 4 6 4 4 7 5 4 ...
## $ trstun : int 5 5 5 4 8 6 6 7 6 8 ...
## $ vote : int 1 1 1 2 1 1 1 1 1 1 ...
## $ prtvtdat : int 1 5 5 66 5 7 1 5 5 2 ...
## $ prtvtebe : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtchr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtccy : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtffi : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtffr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvgde1 : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvgde2 : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtegr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvthhu : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvteis : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvteie : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvteit : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvclt1 : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvclt2 : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvclt3 : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtinl : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtcno : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtfpl : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtept : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtbrs : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtesk : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtgsi : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtges : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtdse : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvthch : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtvtdgb : int NA NA NA NA NA NA NA NA NA NA ...
## $ contplt : int 2 2 1 2 2 2 2 1 2 2 ...
## $ donprty : int 2 2 1 2 2 2 2 2 2 2 ...
## $ badge : int 2 1 2 2 2 2 2 2 2 2 ...
## $ sgnptit : int 2 1 1 2 1 2 2 1 2 2 ...
## $ pbldmna : int 2 1 1 2 2 2 2 1 2 2 ...
## $ bctprd : int 2 1 1 2 2 1 2 1 2 2 ...
## $ pstplonl : int 2 2 1 2 2 2 2 1 2 2 ...
## $ volunfp : int 2 1 1 2 2 2 2 1 2 2 ...
## $ clsprty : int 1 1 1 2 1 2 1 1 2 2 ...
## $ prtcleat : int 1 5 5 66 5 66 1 5 66 66 ...
## $ prtclebe : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclbhr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclccy : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclgfi : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclgfr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclgde : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclegr : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclihu : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtcleis : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclfie : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclfit : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclclt : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclhnl : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclcno : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtcljpl : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclgpt : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclbrs : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclesk : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclgsi : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclhes : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtcldse : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtclhch : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtcldgb : int NA NA NA NA NA NA NA NA NA NA ...
## $ prtdgcl : int 2 2 2 6 2 6 2 2 6 6 ...
## $ lrscale : int 5 0 3 5 2 4 4 3 5 5 ...
## $ stflife : int 8 9 10 7 9 8 8 8 8 8 ...
## $ stfeco : int 6 2 6 4 6 4 6 6 6 7 ...
## $ stfgov : int 4 5 5 4 7 2 3 6 5 8 ...
## $ stfdem : int 6 7 6 6 8 3 6 8 8 7 ...
## $ stfedu : int 8 10 5 5 9 3 88 6 9 8 ...
## $ stfhlth : int 8 10 8 2 9 7 7 7 9 9 ...
## $ gincdif : int 2 1 1 1 2 2 2 2 1 2 ...
## $ freehms : int 2 1 1 2 2 2 2 1 2 2 ...
## [list output truncated]Inspection of hincfel variable
#inspect hincfel variable
str(data$hincfel)## int [1:40156] 1 2 1 2 2 1 2 1 2 2 ...summary(data$hincfel)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.000 1.000 2.000 1.921 2.000 9.000table(data$hincfel)##
## 1 2 3 4 7 8 9
## 14309 18108 5862 1426 299 133 19association_variables <- c("hincfel", "gndr","cntry", "freehms", "stfeco", "stfedu", "prtdgcl","pbldmna","donprty","trstprt")
reduced_data<- data[association_variables]
#conversion of variables into factor type
reduced_data <- reduced_data %>% mutate_if(is.numeric, as.factor)
str(reduced_data)## 'data.frame': 40156 obs. of 10 variables:
## $ hincfel: Factor w/ 7 levels "1","2","3","4",..: 1 2 1 2 2 1 2 1 2 2 ...
## $ gndr : Factor w/ 2 levels "1","2": 1 2 2 2 1 2 2 2 2 2 ...
## $ cntry : chr "AT" "AT" "AT" "AT" ...
## $ freehms: Factor w/ 8 levels "1","2","3","4",..: 2 1 1 2 2 2 2 1 2 2 ...
## $ stfeco : Factor w/ 14 levels "0","1","2","3",..: 7 3 7 5 7 5 7 7 7 8 ...
## $ stfedu : Factor w/ 14 levels "0","1","2","3",..: 9 11 6 6 10 4 13 7 10 9 ...
## $ prtdgcl: Factor w/ 8 levels "1","2","3","4",..: 2 2 2 5 2 5 2 2 5 5 ...
## $ pbldmna: Factor w/ 5 levels "1","2","7","8",..: 2 1 1 2 2 2 2 1 2 2 ...
## $ donprty: Factor w/ 5 levels "1","2","7","8",..: 2 2 1 2 2 2 2 2 2 2 ...
## $ trstprt: Factor w/ 14 levels "0","1","2","3",..: 6 1 5 4 6 6 6 7 3 9 ...summary(reduced_data)## hincfel gndr cntry freehms stfeco
## 1:14309 1:18760 Length:40156 1 :16407 5 :6477
## 2:18108 2:21396 Class :character 2 :13826 6 :5449
## 3: 5862 Mode :character 3 : 4768 3 :5167
## 4: 1426 4 : 2489 4 :5108
## 7: 299 5 : 1905 7 :4899
## 8: 133 8 : 506 2 :3683
## 9: 19 (Other): 255 (Other):9373
## stfedu prtdgcl pbldmna donprty trstprt
## 7 :6211 6 :22766 1: 3296 1: 2448 5 :6710
## 5 :5892 2 : 9984 2:36722 2:37546 0 :5689
## 6 :5272 3 : 3067 7: 63 7: 75 3 :5151
## 8 :4908 1 : 2278 8: 69 8: 80 2 :4642
## 4 :4165 4 : 313 9: 6 9: 7 4 :4555
## 3 :4062 (Other): 154 6 :4245
## (Other):9646 NA's : 1594 (Other):9164reduced_data <- reduced_data %>% mutate_if(is.numeric, as.factor)
#conversion of reduced data into transactions
transactions <- as(reduced_data, "transactions")
# Inspect transactions
summary(transactions)## transactions as itemMatrix in sparse format with
## 40156 rows (elements/itemsets/transactions) and
## 101 columns (items) and a density of 0.09861688
##
## most frequent items:
## donprty=2 pbldmna=2 prtdgcl=6 gndr=2 gndr=1 (Other)
## 37546 36722 22766 21396 18760 262776
##
## element (itemset/transaction) length distribution:
## sizes
## 9 10
## 1594 38562
##
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 9.00 10.00 10.00 9.96 10.00 10.00
##
## includes extended item information - examples:
## labels variables levels
## 1 hincfel=1 hincfel 1
## 2 hincfel=2 hincfel 2
## 3 hincfel=3 hincfel 3
##
## includes extended transaction information - examples:
## transactionID
## 1 1
## 2 2
## 3 3inspect(head(transactions, 5))## items transactionID
## [1] {hincfel=1,
## gndr=1,
## cntry=AT,
## freehms=2,
## stfeco=6,
## stfedu=8,
## prtdgcl=2,
## pbldmna=2,
## donprty=2,
## trstprt=5} 1
## [2] {hincfel=2,
## gndr=2,
## cntry=AT,
## freehms=1,
## stfeco=2,
## stfedu=10,
## prtdgcl=2,
## pbldmna=1,
## donprty=2,
## trstprt=0} 2
## [3] {hincfel=1,
## gndr=2,
## cntry=AT,
## freehms=1,
## stfeco=6,
## stfedu=5,
## prtdgcl=2,
## pbldmna=1,
## donprty=1,
## trstprt=4} 3
## [4] {hincfel=2,
## gndr=2,
## cntry=AT,
## freehms=2,
## stfeco=4,
## stfedu=5,
## prtdgcl=6,
## pbldmna=2,
## donprty=2,
## trstprt=3} 4
## [5] {hincfel=2,
## gndr=1,
## cntry=AT,
## freehms=2,
## stfeco=6,
## stfedu=9,
## prtdgcl=2,
## pbldmna=2,
## donprty=2,
## trstprt=5} 5Sample Image
Image represents a visualization of sampled transactions from the reduced dataset. Rows correspond to observations while columns represent to the factors in the dataset such as hincfel, gndr, cntry.Items with continuous vertical patterns are highly frequent across many transactions. For example, hincfel categories 1 and 2 might dominate the transactions since they are more frequent than others.
image(sample(transactions, 50))
Apriori Algorithm
supp = 0.01: Only rules supported by at least 1% of the transactions are considered. It is used to ensure that rare patterns are excluded from the results.
conf = 0.6: Rules must be correct at least 60% of the time to be included. 0.6 could be considered as a high confidence level which ensures strong associations.
minlen = 2: Each rule must include at least two items.
appearance Argument: Specifies that the rules must include one of the hincfel categories 1, 2, 3, 4, 7, 8, 9 as these are the options for hincfel variable.
Focuses the analysis on how these categories are influenced by other variables.
control = list(verbose = FALSE): Suppresses detailed output during processing.
#Apriori algorithm is applied to find association with hincfel variable with selected columns
rules <- apriori(transactions,
parameter = list(supp = 0.01, conf = 0.6, minlen = 2),
appearance = list(rhs = c("hincfel=1", "hincfel=2", "hincfel=3", "hincfel=4", "hincfel=7", "hincfel=8", "hincfel=9")),
control = list(verbose = FALSE))
rules_sorted <- sort(rules, by = "confidence", decreasing = TRUE)
inspect(head(rules_sorted, 10))## lhs rhs support confidence coverage lift count
## [1] {gndr=1,
## cntry=NL,
## freehms=1,
## pbldmna=2} => {hincfel=1} 0.01003586 0.7647059 0.01312382 2.146029 403
## [2] {gndr=1,
## cntry=NL,
## freehms=1} => {hincfel=1} 0.01038450 0.7623400 0.01362187 2.139390 417
## [3] {cntry=PL,
## prtdgcl=6,
## pbldmna=2,
## donprty=2} => {hincfel=2} 0.01534017 0.7255595 0.02114254 1.608989 616
## [4] {cntry=PL,
## prtdgcl=6,
## pbldmna=2} => {hincfel=2} 0.01576352 0.7250859 0.02174021 1.607938 633
## [5] {cntry=PL,
## freehms=2,
## pbldmna=2,
## donprty=2} => {hincfel=2} 0.01003586 0.7235189 0.01387090 1.604463 403
## [6] {cntry=PL,
## freehms=2,
## pbldmna=2} => {hincfel=2} 0.01038450 0.7227036 0.01436896 1.602656 417
## [7] {cntry=PL,
## prtdgcl=6,
## donprty=2} => {hincfel=2} 0.01556430 0.7175660 0.02169041 1.591262 625
## [8] {cntry=PL,
## prtdgcl=6} => {hincfel=2} 0.01611216 0.7165006 0.02248730 1.588900 647
## [9] {gndr=1,
## cntry=PL,
## pbldmna=2,
## donprty=2} => {hincfel=2} 0.01075804 0.7164179 0.01501644 1.588716 432
## [10] {gndr=1,
## cntry=PL,
## pbldmna=2} => {hincfel=2} 0.01120630 0.7142857 0.01568881 1.583988 450rules_sorted_lift <- sort(rules, by = "lift", decreasing = TRUE)
inspect(head(rules_sorted_lift, 10))## lhs rhs support
## [1] {gndr=1, cntry=NL, freehms=1, pbldmna=2} => {hincfel=1} 0.01003586
## [2] {gndr=1, cntry=NL, freehms=1} => {hincfel=1} 0.01038450
## [3] {gndr=1, cntry=NL, pbldmna=2} => {hincfel=1} 0.01426935
## [4] {cntry=SE, freehms=1, pbldmna=2} => {hincfel=1} 0.01374639
## [5] {gndr=1, cntry=NL} => {hincfel=1} 0.01471760
## [6] {cntry=NL, freehms=1, pbldmna=2} => {hincfel=1} 0.01949895
## [7] {cntry=SE, freehms=1} => {hincfel=1} 0.01484212
## [8] {gndr=1, cntry=NL, pbldmna=2, donprty=2} => {hincfel=1} 0.01277518
## [9] {cntry=NL, freehms=1} => {hincfel=1} 0.02007172
## [10] {gndr=1, cntry=NL, donprty=2} => {hincfel=1} 0.01304911
## confidence coverage lift count
## [1] 0.7647059 0.01312382 2.146029 403
## [2] 0.7623400 0.01362187 2.139390 417
## [3] 0.7047970 0.02024604 1.977904 573
## [4] 0.7031847 0.01954876 1.973379 552
## [5] 0.7010676 0.02099313 1.967438 591
## [6] 0.7003578 0.02784142 1.965446 783
## [7] 0.7003525 0.02119235 1.965431 596
## [8] 0.6970109 0.01832852 1.956053 513
## [9] 0.6966292 0.02881263 1.954982 806
## [10] 0.6949602 0.01877677 1.950299 524rules_sorted_support <- sort(rules, by = "support", decreasing = TRUE)
inspect(head(rules_sorted_support, 10))## lhs rhs support confidence coverage lift count
## [1] {freehms=1,
## stfeco=7} => {hincfel=1} 0.03309593 0.6051913 0.05468672 1.698376 1329
## [2] {cntry=HU} => {hincfel=2} 0.03185078 0.6038716 0.05274430 1.339136 1279
## [3] {cntry=HU,
## donprty=2} => {hincfel=2} 0.03068035 0.6036257 0.05082678 1.338590 1232
## [4] {cntry=HU,
## pbldmna=2} => {hincfel=2} 0.03058074 0.6031434 0.05070226 1.337521 1228
## [5] {freehms=1,
## stfeco=7,
## pbldmna=2} => {hincfel=1} 0.02993326 0.6092245 0.04913338 1.709695 1202
## [6] {cntry=HU,
## pbldmna=2,
## donprty=2} => {hincfel=2} 0.02963443 0.6028369 0.04915828 1.336841 1190
## [7] {cntry=NL} => {hincfel=1} 0.02749278 0.6513274 0.04221038 1.827850 1104
## [8] {freehms=1,
## stfeco=7,
## pbldmna=2,
## donprty=2} => {hincfel=1} 0.02724375 0.6040861 0.04509911 1.695275 1094
## [9] {cntry=NL,
## pbldmna=2} => {hincfel=1} 0.02674569 0.6548780 0.04084072 1.837814 1074
## [10] {cntry=PL} => {hincfel=2} 0.02520171 0.7018031 0.03590995 1.556307 1012Visualizations
This graph represents the relationships between items (variables and categories) in the association rules generated by the Apriori algorithm.
Nodes: Represent items such as hincfel, gndr. Larger nodes indicate items with higher support which means they are more frequent in the dataset.
Edges: Represent the associations or rules between items. Thicker or darker edges imply stronger relationships which means higher confidence or lift.
Clusters: Nodes are grouped into clusters based on their relationships. These clusters represent groups of items that frequently occurs together.
Color Intensity: The color of the nodes indicates the strength of the association. Darker color implies stronger association
plot(rules_sorted, method = "graph", engine = "htmlwidget")Scatter Plot
This plot illustrates the rules in terms of their support (x-axis), confidence (y-axis), and lift (color intensity).
Most rules have support in the range of 0.01–0.03 and confidence between 0.6–0.76. Rules with higher lift are shaded darker, indicating stronger associations. Few rules stand out with very high confidence (~0.76) but lower support, suggesting these rules are reliable but not common.
plot(rules_sorted, measure = c("support", "confidence"), shading = "lift", engine = "plotly")Parallel Coordinates Plot
This plot shows how the LHS and RHS are connected across multiple rules.
The overlapping red lines indicate common patterns where antecedents contribute to multiple hincfel outcomes.
The far-right axis (rhs) represents the target variable hincfel.Each line represents a rule.The path shows how items on the left axes connect to a specific hincfel category on the rhs.Darker red lines indicate stronger rules like higher confidence or lift. Lighter colored lines represent weaker rules. Thicker lines suggest rules with higher support.
Most of the strong rules (dark red lines) end at hincfel=1 or hincfel=2. Moreover, variables like stfeco=7, stfedu=7, cntry=NL, and freehms=2 frequently appear, which means they are strongly associated with specific hincfel categories. On the other hand, certain antecedents are commonly grouped together to predict specific outcomes. For example, combinations of stfeco=7, prtdgcl=2, and donprty=1 frequently lead to hincfel=1.
plot(rules_sorted, method = "paracoord", control = list(reorder = TRUE))
Grouped Matrix Plot
This plot groups rules based on their antecedents and shows their relationships to consequents.
The size of the circles represents support, while the color represents lift. Strong rules with high lift and support are observed for hincfel=1 and hincfel=2.
Most rules are associated with hincfel=1, as indicated by the more frequent and larger circles in the first row. This aligns with hincfel=1 being one of the most frequent categories in the dataset. Fewer and smaller circles are observed for hincfel=2, reflecting lower support for rules leading to this category.
Variables like satisfaction with the economy (stfeco=7) and satisfaction with education (stfedu=7) are key factors driving positive perceptions of household income (hincfel=1). Furthermore, groups like {stfeco=7, stfedu=7} frequently influence hincfel=1.
plot(rules_sorted, method = "grouped", control = list(k = 15))
Matrix Plot
plot(rules_sorted, method = "matrix", control = list(reorder = "none"))## Itemsets in Antecedent (LHS)
## [1] "{gndr=1,cntry=NL,freehms=1,pbldmna=2}"
## [2] "{gndr=1,cntry=NL,freehms=1}"
## [3] "{cntry=PL,prtdgcl=6,pbldmna=2,donprty=2}"
## [4] "{cntry=PL,prtdgcl=6,pbldmna=2}"
## [5] "{cntry=PL,freehms=2,pbldmna=2,donprty=2}"
## [6] "{cntry=PL,freehms=2,pbldmna=2}"
## [7] "{cntry=PL,prtdgcl=6,donprty=2}"
## [8] "{cntry=PL,prtdgcl=6}"
## [9] "{gndr=1,cntry=PL,pbldmna=2,donprty=2}"
## [10] "{gndr=1,cntry=PL,pbldmna=2}"
## [11] "{cntry=PL,freehms=2}"
## [12] "{cntry=PL,freehms=2,donprty=2}"
## [13] "{cntry=PL,pbldmna=2,donprty=2}"
## [14] "{cntry=PL,pbldmna=2}"
## [15] "{gndr=2,cntry=PL,pbldmna=2}"
## [16] "{gndr=2,cntry=PL,pbldmna=2,donprty=2}"
## [17] "{gndr=1,cntry=PL,donprty=2}"
## [18] "{gndr=1,cntry=NL,pbldmna=2}"
## [19] "{gndr=1,cntry=PL}"
## [20] "{cntry=PL,donprty=2}"
## [21] "{cntry=SE,freehms=1,pbldmna=2}"
## [22] "{cntry=PL}"
## [23] "{gndr=1,cntry=NL}"
## [24] "{cntry=NL,freehms=1,pbldmna=2}"
## [25] "{cntry=SE,freehms=1}"
## [26] "{gndr=2,cntry=PL}"
## [27] "{gndr=2,cntry=PL,donprty=2}"
## [28] "{gndr=1,cntry=NL,pbldmna=2,donprty=2}"
## [29] "{cntry=NL,freehms=1}"
## [30] "{gndr=1,cntry=NL,donprty=2}"
## [31] "{cntry=NL,freehms=1,pbldmna=2,donprty=2}"
## [32] "{cntry=SE,freehms=1,donprty=2}"
## [33] "{cntry=SE,freehms=1,pbldmna=2,donprty=2}"
## [34] "{cntry=NL,freehms=1,donprty=2}"
## [35] "{gndr=1,cntry=SE}"
## [36] "{cntry=NL,prtdgcl=2,pbldmna=2}"
## [37] "{cntry=NL,prtdgcl=2}"
## [38] "{gndr=1,cntry=SE,pbldmna=2}"
## [39] "{gndr=1,freehms=1,pbldmna=2,trstprt=7}"
## [40] "{cntry=SE,pbldmna=2}"
## [41] "{cntry=SE}"
## [42] "{gndr=1,freehms=1,stfeco=8,pbldmna=2}"
## [43] "{gndr=1,freehms=1,trstprt=7}"
## [44] "{cntry=SE,donprty=2}"
## [45] "{cntry=SE,pbldmna=2,donprty=2}"
## [46] "{gndr=1,freehms=1,stfeco=8}"
## [47] "{cntry=NL,pbldmna=2}"
## [48] "{gndr=1,freehms=1,stfeco=7,pbldmna=2}"
## [49] "{cntry=IS,pbldmna=2}"
## [50] "{cntry=NL}"
## [51] "{gndr=1,freehms=1,stfeco=7,pbldmna=2,donprty=2}"
## [52] "{freehms=1,prtdgcl=2,trstprt=7}"
## [53] "{gndr=1,freehms=1,stfeco=7}"
## [54] "{cntry=NL,pbldmna=2,donprty=2}"
## [55] "{cntry=NL,donprty=2}"
## [56] "{gndr=1,freehms=1,pbldmna=2,trstprt=6}"
## [57] "{gndr=1,freehms=1,stfeco=7,donprty=2}"
## [58] "{cntry=AT,freehms=2,prtdgcl=6,pbldmna=2,donprty=2}"
## [59] "{cntry=AT,freehms=2,prtdgcl=6,donprty=2}"
## [60] "{freehms=1,stfeco=7,prtdgcl=2,pbldmna=2}"
## [61] "{freehms=1,stfeco=8,pbldmna=2}"
## [62] "{cntry=AT,freehms=2,prtdgcl=6,pbldmna=2}"
## [63] "{cntry=AT,freehms=2,prtdgcl=6}"
## [64] "{gndr=1,freehms=1,pbldmna=2,donprty=2,trstprt=6}"
## [65] "{freehms=1,stfeco=7,prtdgcl=2}"
## [66] "{gndr=1,freehms=1,trstprt=6}"
## [67] "{cntry=IS}"
## [68] "{freehms=1,pbldmna=2,donprty=1}"
## [69] "{cntry=NL,prtdgcl=6,pbldmna=2,donprty=2}"
## [70] "{gndr=2,cntry=HU,prtdgcl=6,pbldmna=2,donprty=2}"
## [71] "{freehms=1,stfeco=8}"
## [72] "{freehms=1,stfeco=8,pbldmna=2,donprty=2}"
## [73] "{cntry=NL,prtdgcl=6,pbldmna=2}"
## [74] "{gndr=2,cntry=HU,prtdgcl=6,pbldmna=2}"
## [75] "{gndr=2,cntry=HU,prtdgcl=6,donprty=2}"
## [76] "{gndr=1,freehms=1,donprty=1}"
## [77] "{cntry=IS,donprty=2}"
## [78] "{cntry=NL,prtdgcl=6,donprty=2}"
## [79] "{gndr=2,cntry=HU,prtdgcl=6}"
## [80] "{gndr=1,freehms=1,donprty=2,trstprt=6}"
## [81] "{freehms=1,pbldmna=2,trstprt=7}"
## [82] "{freehms=1,stfeco=8,donprty=2}"
## [83] "{cntry=NL,prtdgcl=6}"
## [84] "{freehms=1,trstprt=7}"
## [85] "{cntry=HU,prtdgcl=6,pbldmna=2,donprty=2}"
## [86] "{freehms=1,prtdgcl=2,pbldmna=2,trstprt=6}"
## [87] "{cntry=HU,prtdgcl=6,pbldmna=2}"
## [88] "{cntry=HU,prtdgcl=6,donprty=2}"
## [89] "{gndr=2,cntry=HU,pbldmna=2,donprty=2}"
## [90] "{gndr=2,cntry=HU,pbldmna=2}"
## [91] "{gndr=2,cntry=HU,donprty=2}"
## [92] "{cntry=HU,prtdgcl=6}"
## [93] "{gndr=2,cntry=HU}"
## [94] "{freehms=1,prtdgcl=2,trstprt=6}"
## [95] "{freehms=1,stfeco=7,pbldmna=2}"
## [96] "{freehms=1,pbldmna=2,donprty=2,trstprt=7}"
## [97] "{gndr=2,cntry=NL,pbldmna=2}"
## [98] "{freehms=1,stfedu=7,prtdgcl=2,pbldmna=2}"
## [99] "{freehms=1,stfeco=7}"
## [100] "{gndr=1,freehms=1,stfedu=7,pbldmna=2}"
## [101] "{freehms=1,stfeco=7,pbldmna=2,donprty=2}"
## [102] "{cntry=HU}"
## [103] "{cntry=HU,donprty=2}"
## [104] "{cntry=HU,pbldmna=2}"
## [105] "{cntry=HU,pbldmna=2,donprty=2}"
## [106] "{gndr=2,cntry=NL}"
## [107] "{gndr=1,freehms=1,stfedu=7}"
## [108] "{cntry=AT,freehms=2,pbldmna=2,donprty=2}"
## [109] "{prtdgcl=2,pbldmna=2,donprty=1}"
## Itemsets in Consequent (RHS)
## [1] "{hincfel=2}" "{hincfel=1}"
This matrix visualization provides a heatmap-style representation of association rules, highlighting the relationships between the left hand side and right hand side of the rules.
Rows: Each row represents a category of the target variable as hincfel=1, hincfel=2.
Columns: Each column represents a unique set of antecedents (conditions) on the LHS of the rules. Darker columns indicate strong associations with high lift.
Color Intensity: The color represents the lift of the rule, where dark red implies higher lift (stronger association) and light red means lower lift (weaker association).
The darkest red sections in the first row indicate the strongest rules for hincfel=1.Although hincfel=2 also shows strong associations, they are generally less frequent compared to hincfel=1