knitr::opts_chunk$set(echo = TRUE)

Data Science Jobs in India: From Import to Visualization

Project Information

  • Project Title: Descriptive Analytics of Data Science Jobs in India
  • Date: 2025-10-30
  • Student Name: Sudhanshu Ranjan
  • Student Registration: 12313589
  • Student Roll No: 66
  • Student Group: 2
  • Section: D2302

Introduction

This project analyzes trends and statistics in Data Science jobs in India. We use descriptive analytics and visualization to explore job titles, companies, experience required, and salary patterns.

Load Required Libraries

library(dplyr)
library(ggplot2)
library(class)
library(cluster)

Import & Clean the Dataset

# Set working directory
setwd("C:/Users/sudha/Downloads/archive")  # Adjust to your actual folder

# Load the data
dsjobs <- read.csv("Data_Science_Jobs_in_India.csv")

# Convert salary columns to numeric (remove 'L')
dsjobs$avg_salary_num <- as.numeric(gsub("L", "", dsjobs$avg_salary))
dsjobs$min_salary_num <- as.numeric(gsub("L", "", dsjobs$min_salary))
dsjobs$max_salary_num <- as.numeric(gsub("L", "", dsjobs$max_salary))

Dataset Overview

str(dsjobs)
summary(dsjobs)
head(dsjobs)
cat("Number of Rows:", nrow(dsjobs), "\n")
cat("Number of Columns:", ncol(dsjobs), "\n")

Descriptive Statistics

# Mean, Median, Standard Deviation
mean(dsjobs$avg_salary_num, na.rm = TRUE)
median(dsjobs$avg_salary_num, na.rm = TRUE)
sd(dsjobs$avg_salary_num, na.rm = TRUE)

# Mode function for salary
getmode <- function(v) { uniqv <- unique(v); uniqv[which.max(tabulate(match(v, uniqv)))] }
getmode(dsjobs$avg_salary)

Grouped Statistics by Job Title

dsjobs %>%
  group_by(job_title) %>%
  summarise(
    Mean_Salary = mean(avg_salary_num, na.rm = TRUE),
    Mean_Exp = mean(min_experience, na.rm = TRUE),
    Count = n()
  )

Correlation Matrix

cor(dsjobs[, c("avg_salary_num", "min_salary_num", "max_salary_num", "min_experience")])

Linear Regression: Salary vs Experience

lm_model <- lm(avg_salary_num ~ min_experience, data = dsjobs)
summary(lm_model)
plot(dsjobs$min_experience, dsjobs$avg_salary_num, main = "Salary vs Experience")
abline(lm_model, col = "red")

ANOVA: Salary by Job Title

anova_model <- aov(avg_salary_num ~ job_title, data = dsjobs)
summary(anova_model)
boxplot(avg_salary_num ~ job_title, data = dsjobs, las = 2, main = "Salary by Job")

Clustering (K-Means)

set.seed(101)
cluster_data <- dsjobs[, c("avg_salary_num", "min_experience")]
fit <- kmeans(cluster_data, centers = 3)
dsjobs$cluster <- fit$cluster
ggplot(dsjobs, aes(x = min_experience, y = avg_salary_num, color = factor(cluster))) +
  geom_point() +
  labs(title = "K-Means Clustering: Salary & Experience")

Classification (KNN)

set.seed(100)
split <- sample(1:nrow(dsjobs), size = 0.7 * nrow(dsjobs))
train <- dsjobs[split, ]
test <- dsjobs[-split, ]
knn_pred <- knn(train = train[, c("avg_salary_num", "min_experience")],
                test = test[, c("avg_salary_num", "min_experience")],
                cl = train$job_title, k = 3)
table(knn_pred, test$job_title)

Data Visualization

# Histogram
ggplot(dsjobs, aes(x = avg_salary_num, fill = job_title)) +
  geom_histogram(bins = 20, color = "black", alpha = 0.7) +
  labs(title = "Distribution of Salary by Job Title", x = "Average Salary (L)", y = "Count") +
  theme_minimal()

# Boxplot
ggplot(dsjobs, aes(x = job_title, y = avg_salary_num, fill = job_title)) +
  geom_boxplot(alpha = 0.8) +
  labs(title = "Salary by Job Title", x = "Job Title", y = "Average Salary (L)") +
  theme_minimal()

# Scatter Plot
ggplot(dsjobs, aes(x = min_experience, y = avg_salary_num, color = job_title)) +
  geom_point(size = 3, alpha = 0.8) +
  labs(title = "Experience vs Salary", x = "Minimum Experience", y = "Average Salary (L)") +
  theme_minimal()

# Pair Plot
pairs(dsjobs[, c("avg_salary_num", "min_salary_num", "max_salary_num", "min_experience")], main = "Pair Plot of Numeric Features")

# Density Plot
ggplot(dsjobs, aes(x = avg_salary_num, fill = job_title)) +
  geom_density(alpha = 0.6) +
  labs(title = "Density Plot of Salary by Job Title", x = "Average Salary (L)", y = "Density") +
  theme_minimal()

Insights and Interpretation

  • Higher experience often means higher average salary.
  • Job title affects salary significantly.
  • Clustering reveals groups of similar jobs.
  • Visualizations show data range, outliers, and trends.

Conclusion

This workflow demonstrates complete descriptive analytics for Data Science jobs in India using clean, numeric data and clear, reproducible R code. ```

How to use: - Save as YourProject.Rmd and open in RStudio. - Knit to HTML—the code and outputs will show together for every block. - Update file path if needed and check your dataset filename.

Now your code, output, and plots will display, and all variables/columns are correct for your specific CSV file!