• Start date: September, 2024

R tutorial for beginners 2022; r programming full course in 7 hours- r tutorial simplilearn.

• Variables in R
• Data types in R
• Logical operators
• Print formatting
• Vectors

v1 <- c(1,2,3,4,5)
V2 <- c("red", "green", "yellow")
V2

## [1] "red"    "green"  "yellow"

• List

List_1 <- list(x=c(10,20,30),
Y=c("a", "b", "c"),
Z=c(TRUE, FALSE))
List_1

## $x
## [1] 10 20 30
## 
## $Y
## [1] "a" "b" "c"
## 
## $Z
## [1]  TRUE FALSE

• Matrix

M <- matrix(1:9, nrow=3,
ncol=3, byrow=T)
M

##      [,1] [,2] [,3]
## [1,]    1    2    3
## [2,]    4    5    6
## [3,]    7    8    9

• Data frame

BMI <- data.frame(
Gender = c("Male", "Male", "Female"),
Height = c(152, 171.5, 165),
Weight = c(81, 93, 78),
Age = c(42, 38,26))

BMI

##   Gender Height Weight Age
## 1   Male  152.0     81  42
## 2   Male  171.5     93  38
## 3 Female  165.0     78  26

• If statement in R

X <- 30L 
  if(is.integer(X)) {
    print("x is an Integer")
}

## [1] "x is an Integer"

• While and for loop in R

V <- c("Hello World")
        count <- 2
        while (count < 7){
print(V)
count = count + 3}

## [1] "Hello World"
## [1] "Hello World"

fruit <- c("Apple", "Orange", "Passion fruit", "Banana")
    for (i in fruit){
print(i)}

## [1] "Apple"
## [1] "Orange"
## [1] "Passion fruit"
## [1] "Banana"

• Functions in R

I <- function(S,a,b,C){
  C-S+(a/b)*log(S)
}

• Regular expressions in R used for pattern matching{grepl() and grep()}
• Built-in R functions
  • seq(); creating sequences
  • sort(); rearranging sequences
  • rev(); reversing elements in R objects
  • append(); combining objects in R
• Data manipulation in R;dplyr, tidyr.
  • etc

• Representing data; including how to input data in R, where to place this data, how to prepare data from further analysis.
• Transforming data; which includes adjusting the arrangement or view of the data-set.
• Applying functions; applying a formula or an equation to an object using “function” so experiments can be carried out by only calling the object that is now a function.
• Tidying Data; involves the cleaning and preparation process of a data-set before further analysis.
• Visualizing Data; involves creating visual representation of the data to allow for easy understanding and explanation.

• Data visualization

library(tidyverse)
library(dplyr)
library(ggplot2)

fouye <- read.csv("C:\\Users\\hp\\Desktop\\RRR\\FUOYE ENROLLMENT.csv")
fouye <- subset(fouye, !(TOTAL.ENROLLED == "NA"))

fouye_plots_social <- fouye[,(c(1,14))]
fou14 <- ggplot(fouye_plots_social, aes(x=YEAR, y=SOCIAL.SCIENCES))+
  geom_bar(stat = "identity", aes(fill = SOCIAL.SCIENCES))+
  theme_minimal()+
  scale_x_continuous(limits=c(2010,2025))

fou14

• Growth rate(and visualization)

fouye_plots_total <- fouye[,(c(1,15))]
fouye_total_growth <- fouye_plots_total|>
  arrange(YEAR)|>
  mutate(growth_rate=(TOTAL.ENROLLED/lag(TOTAL.ENROLLED)-1)*100)
head(fouye_total_growth, 3)

##   YEAR TOTAL.ENROLLED growth_rate
## 1 2011            423          NA
## 2 2012           1144   170.44917
## 3 2013           1480    29.37063

• Future prediction with lower and upper limits(linear models with visualization)

fouye_model <- lm(TOTAL.ENROLLED~ YEAR, data = fouye_plots_total)

new_fouye_years <- data.frame(YEAR=c(2024,2025,2026,2027,2028,2029,2030,2031,
                                     2032,2033,2034,2035))

predictions <- predict(fouye_model, newdata = new_fouye_years)

model_results <- data.frame(YEAR=new_fouye_years$YEAR,
                            TOTAL.ENROLLED=predictions
                            )
head(model_results, 5)

##   YEAR TOTAL.ENROLLED
## 1 2024       41428.14
## 2 2025       45075.27
## 3 2026       48722.39
## 4 2027       52369.51
## 5 2028       56016.63

The growth rate plot

growth_rater <- ggplot(fouye_total_growth)+
  #bar chart
  geom_bar(aes(x=YEAR,y=TOTAL.ENROLLED),stat = "identity", fill="steelblue")+
  #line graph for growth
  geom_line(aes(x=YEAR, y=growth_rate*max(TOTAL.ENROLLED)/max(growth_rate, na.rm = TRUE)),
            color="red", size=1,group=1)+
  geom_point(aes(x=YEAR,y=growth_rate*max(TOTAL.ENROLLED)/max(growth_rate, na.rm = TRUE)),
             color="darkred", size=3)+
  scale_y_continuous(name = "TOTAL.ENROLLED",
                     #SECONDARY AXIS
                     sec.axis = sec_axis(~.*max(fouye_total_growth$growth_rate,na.rm = T)/
                                           max(fouye_total_growth$TOTAL.ENROLLED),
                                         name = "Growth Rate (%)"))+
  theme_minimal()+
  theme(
    axis.title.y.left = element_text(color = "steelblue"),
    axis.title.y.right = element_text(color = "red")
  )+labs(title = "Enrollment Over Time With Growth Rate", x="Year")+
  scale_x_continuous(limits = c(2010,2025))

growth_rater

• Data manipulation; stringr, dplyr, tidyr
  • Selecting columns needed for specific tasks

cognify <- read.csv("C:\\Users\\hp\\Desktop\\RRR\\cognify.csv")

city_rate <- cognify[,as.double(c(4,18))]
head(city_rate, 3)

##               City Aggregate.rating
## 1      Makati City              4.8
## 2      Makati City              4.5
## 3 Mandaluyong City              4.4

For tasks on that average rating for each city, only two columns were needed; the city and aggregate rating columns

• Comparison based on a column of TRUE or FALSE

online_del <- cognify[,c(14,18)]
total_restau <- nrow(cognify)
percent_online <-   sum     (((online_del$Has.Online.delivery=="Yes")
                          /total_restau)*100)
percent_online

## [1] 25.66223

• Analysis of cuisine names and determining combinations

agg_cognify <- cognify|>
  group_by(Cuisines)|>
  summarise(
    avg_rating=mean(Aggregate.rating, na.rm=TRUE),
    count=n())|>
  arrange(desc(avg_rating))

head(agg_cognify, 2)

## # A tibble: 2 × 3
##   Cuisines                avg_rating count
##   <chr>                        <dbl> <int>
## 1 American, BBQ, Sandwich        4.9     1
## 2 American, Burger, Grill        4.9     1

• Analysis of restaurant names and determining chains

chains_res <- cognify[,c(2,18,21)]
chainss <- chains_res|>group_by(Restaurant.Name)|>
        summarise(no=n())|>arrange(desc(no))
head(chainss, 2)

## # A tibble: 2 × 2
##   Restaurant.Name    no
##   <chr>           <int>
## 1 Cafe Coffee Day    83
## 2 Domino's Pizza     79

• Review text analysis

rate_text <- cognify[,c(2,18,20)]
rate_count <- rate_text|>
        group_by(Rating.text)|>
        summarise(no=n())
rate_count

## # A tibble: 6 × 2
##   Rating.text    no
##   <chr>       <int>
## 1 Average      3737
## 2 Excellent     301
## 3 Good         2100
## 4 Not rated    2148
## 5 Poor          186
## 6 Very Good    1079

rate_text <- rate_text|>
            mutate(Review_length=nchar(Rating.text))

head(rate_text, 3)

##          Restaurant.Name Aggregate.rating Rating.text Review_length
## 1       Le Petit Souffle              4.8   Excellent             9
## 2       Izakaya Kikufuji              4.5   Excellent             9
## 3 Heat - Edsa Shangri-La              4.4   Very Good             9

• Data Visualization; ggplot2, leaflet(maps)
  • Histogram to show price distribution of restaurants

price_range <- cognify[, as.numeric(c(2,17))]
price_range <- price_range|>
  group_by(Price.range)|>
  arrange(desc(Price.range))
pr <- ggplot(price_range, aes(x=Price.range))+
  geom_histogram(fill="darkorange")+
  theme_minimal()

pr

• Interactive world map to show location of restaurants, with names and rating plus clustering

cognify_map <- cognify[,c(2,8,9,18)]
world <- map_data("world")

map1 <- ggplot()+
  geom_polygon(data = world, aes(x=long, y=lat, group = group),
               fill="lightblue", color="darkblue")+
  geom_point(data=cognify_map, aes(x=Longitude, y=Latitude,
                                   fill=Aggregate.rating,
                                   colour = "darkred"), size=3)+
  geom_text(data = cognify_map, aes(x=Longitude, y=Latitude,
                                    label = Restaurant.Name),
            vjust=-1, hjust=0.5, size=3, fontface="bold")+
  coord_fixed(1.3)+
  labs(title = "Restaurant Locations", x="Longitude", y="Latitude")+
  theme_minimal()

map1

The ggplot2 map could not deal with the clustering of the restaurant names on the map which caused pivoting to the “leaflet” package

library(leaflet)
interact_map <- leaflet(cognify_map) %>%
    addTiles() %>%
    addMarkers(
            lng = ~Longitude, 
            lat = ~Latitude, 
            popup = ~paste(
         "<b>Restaurant:</b>",Restaurant.Name,"<br>",
          "<b>Rating:</b>",Aggregate.rating,"<br>"
          ),  # Show restaurant name on click
    clusterOptions = markerClusterOptions()  # Enables clustering
  )

interact_map

• Chisq.test
  • Determining if the price ranges are affected by different factors e.g Online delivery

online_dev <- cognify[,c(2,13,14,17)]
chisq.test(table(online_dev$Price.range, online_dev$Has.Online.delivery))

## 
##  Pearson's Chi-squared test
## 
## data:  table(online_dev$Price.range, online_dev$Has.Online.delivery)
## X-squared = 721.38, df = 3, p-value < 2.2e-16

Certificate earned

• Data visualizations and details needed for the executives of a business/company

library(dplyr)
library(stringr)
library(ggplot2)
library(anytime)
library(lubridate)
library(scales)
library(tidyr)
tata <- read.csv("C:\\Users\\hp\\Desktop\\RRR\\Online Retail.csv")

tata$Quantity <- as.numeric(tata$Quantity)
tata$UnitPrice <- as.numeric(tata$UnitPrice)
tata$Quantity[tata$Quantity < 1] <- NA
tata$UnitPrice[tata$UnitPrice < 0] <- NA

no_uk <- tata|>
  filter(Country != "United Kingdom")
no_uk <- no_uk[,c(4,6,8)]

no_uk_summary <- no_uk |>
  group_by(Country) |>
  summarise(
    Revenue = sum(Quantity*UnitPrice, na.rm = TRUE)
  )|>
  filter(Revenue!=0)

no_uk_quan <- no_uk[,c(1,3)]

no_uk_quan <- no_uk_quan|>
  group_by(Country)|>
  summarise(
    Quantities= sum(Quantity, na.rm = TRUE)
  )|>arrange(Country)|>
  filter(Quantities!=0)|>
  filter(Country!="")

no_uk_bar <- merge(no_uk_quan,no_uk_summary, by="Country")
no_uk_bar <- no_uk_bar|>
  arrange(desc(Revenue))
no_uk_bar_long <- pivot_longer(no_uk_bar,
  cols = c(Quantities,Revenue),
  names_to = "variable",
  values_to = "value"
)

ggplot(no_uk_bar_long, aes(x=reorder(Country,value), y=value, fill = variable))+
  geom_bar(stat = "identity", position="dodge", width = 0.7)+
  theme_minimal()+
  labs(title = "Side-by-side bar Plot",
       x="Country", y="Revenue", fill="Quantity")+
  theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5))

• Topic(s) to cover
  • Linear regression as shallow neural network
  • Logistic regression as a neural network
  • Artificial Neural Network (ANN)
  • Deep Neural Network
  • Bias and Variance in deep learning, etc.
• Expected outcome
  • Basic understanding of Artificial intelligence with R

Finding relationships and uses of Artificial Intelligence with Data analysis and of course the application of these in the industrial chemistry world