Code of all the analysis

Author: emu

data_visualization.R

@@ -0,0 +1,29 @@
+library(ggplot2)
+
+ggplot(data = mtcars, aes(x = am)) + 
+       geom_bar() + 
+  ggtitle("Car count based on Transmission Type [0 = manual, 1 = auto]") +
+  xlab("Transmission Type") +
+  ylab("Total cars")
+      
+       
+ggplot(data = mtcars, aes(x = qsec)) + 
+  geom_histogram(bins = 10) + 
+  ggtitle("Distribution of Fuel") +
+  xlab("Fuel per Milage") +
+  ylab("Total cars")
+
+ggplot(data = mtcars, aes(x = qsec)) + 
+  geom_density() + 
+  ggtitle("Distribution of Fuel") +
+  xlab("Fuel per Milage") +
+  ylab("Total cars")
+
+ggplot(data = mtcars, aes(x = cyl, y = qsec)) + 
+  geom_point() + 
+  ggtitle("Distribution of Fuel") +
+  xlab("cylinders") +
+  ylab("Total milage")
+
+
+

descriptive_statistics.R

@@ -0,0 +1,30 @@
+mtcars
+
+# find number of cars based on Automatic/Manual gear
+table(mtcars$am)
+
+# minimum qsec value
+min(mtcars$qsec)
+
+# maximum qsec value
+max(mtcars$qsec)
+
+# mean qsec value
+mean(mtcars$qsec)
+
+# median qsec value
+median(mtcars$qsec)
+
+# quartile qsec value
+quantile(mtcars$qsec)
+
+# sum of qsec value
+sum(mtcars$qsec)
+
+# correclation coefficients between cyl and qsec
+cor(
+  x = mtcars$cyl,
+  y = mtcars$qsec
+)
+
+# summary(mtcars)

intro_demo.R

@@ -0,0 +1,8 @@
+gender <- c('Male', 'Female', 'Female', 'Male', 'Female', 'Male')
+gender <- factor(gender)
+unclass(gender)
+
+genders <- c('Male', 'Female')
+people <- c('Kabir', 'Laif', 'Moin')
+df <- data.frame(colnames(genders), people)
+df

predict_with_ml.R

@@ -0,0 +1,74 @@
+library(tree)
+library(ggplot2)
+library(RColorBrewer)
+library(caret)
+
+data("iris")
+
+# set a seed to make randomness reproducable
+set.seed(42)
+
+# get 100 random indexes between 1 to 150
+indexes <- sample(
+  x = 1:150,
+  size = 100
+)
+
+print(indexes)
+
+# based on the indexes split iris data to TRAIN & TEST
+train <- iris[indexes, ]
+test <- iris[-indexes, ]
+
+# create a decision trre model based on train data
+model <- tree(
+  data = train,
+  formula = Species ~ . 
+)
+
+# get summary of the created model
+summary(model)
+
+# plot tree with text
+plot(model)
+text(model)
+
+# create a color palatte
+palatte <- brewer.pal(3, "Set2")
+
+# create a scatterplot with colored species
+plot(
+  x = iris$Petal.Length,
+  y = iris$Petal.Width,
+  pch = 19,
+  col = palatte[as.numeric(iris$Species)],
+  main = "IRIS petal Length vs Width",
+  xlab = "Petal Length",
+  ylab = "Petal Width"
+)
+
+# partition based on species
+partition.tree(
+  tree = model,
+  label = "Species",
+  add = TRUE
+)
+
+# Predict test data 
+predictions <- predict(
+  object = model,
+  newdata = test,
+  type = "class"
+)
+
+# create a confussion matrix
+table(
+  x = predictions,
+  y = test$Species
+)
+
+# evaluate the predicting result 
+confusionMatrix(
+  data = predictions,
+  reference = test$Species
+)

statistical_models.R

@@ -0,0 +1,35 @@
+library(ggplot2)
+
+data("iris")
+
+head(iris)
+
+# draw a scatterplot based on petal width
+petal_plot <- ggplot(data = iris, aes(x = Petal.Width, y = Petal.Length)) +
+              geom_point() + 
+              ggtitle("Scatterplot of petal width & length") +
+              xlab("petal width") +
+              ylab("petal length")
+
+# create a linear regression model
+linear_model <- lm(
+  formula = Petal.Width ~ Petal.Length,
+  data = iris
+)
+
+# summary(linear_model)
+
+# draw linear regression line
+petal_plot + stat_smooth(method = "lm", col = "red")
+
+# getting correlation coefficients
+cor(
+  x = iris$Petal.Width,
+  y = iris$Petal.Length
+)
+
+# predict new values from the model
+predict(
+  object = linear_model,
+  newdata = data.frame(Petal.Length = c(2,5,7))
+)