03 — Projects
Selected work
End-to-end analyses — from raw transactional data to dashboards ready for a stakeholder review. Click a card for details.
Engineered features across 3,900 transactions — imputing missing ratings and creating age-group and purchase-frequency fields to power a customer segmentation dashboard.
3,900transactions
2.1×revenue, male vs. female customers
73%revenue from non-subscribers
My contribution
- — Imputed 37 missing ratings and engineered age-group / purchase-frequency fields for behavior analysis.
- — Ran 10 SQL analyses against MySQL to surface revenue and subscription patterns.
- — Designed a Power BI segmentation dashboard grouping customers into Loyal (3,116), Returning (701), and New (83) tiers for targeted retention and marketing.
HR Analytics — Employee Attrition Analysis
Power BIDAXPower QueryExcel
View on GitHub ↗
Transformed a 1,470-employee, 35+ attribute HR dataset to build an interactive attrition dashboard, surfacing where and why employees were leaving.
1,470employees analyzed
16%overall attrition rate
26–35highest-attrition age band
My contribution
- — Cleaned the dataset in Power Query — removing duplicates, standardizing fields, and building calculated columns for attrition analysis.
- — Built an interactive Power BI dashboard using DAX, KPI cards, and slicers.
- — Surfaced salary-band, education-field, and job-satisfaction patterns to inform workforce retention, with attrition highest among employees with under 2 years of tenure.
E-Commerce Customer Behavior & Sales Analytics
PythonSQLMySQLPower BIPower Query
View on GitHub ↗
Analyzed 17,049 transactions from 5,000 customers with Python and SQL, then built a DAX-driven RFM segmentation dashboard in Power BI.
₹21.78Mtotal sales analyzed
88.2%returning-customer transactions
922Champions segment customers
My contribution
- — Prepared and explored 17,049 transactions from 5,000 customers using Python and Pandas.
- — Applied CTEs, CASE, RANK, DENSE RANK, and window functions in SQL to identify sales, customer, product, and trend patterns.
- — Implemented RFM segmentation in Power BI, identifying 922 Champions and 1,346 Loyal Customers.
Built an interactive Power BI dashboard to analyze mobile sales across brands, cities, payment methods, customer ratings, and monthly sales trends.
769Mtotal sales
19Kquantity sold
3,835transactions
My contribution
- — Developed KPI cards for total sales, quantity sold, transactions, and average price per unit.
- — Created visualizations for brand performance, payment methods, customer ratings, monthly sales, and city-wise sales.
- — Added interactive filters for month, mobile brand, and payment method using Power BI.
Customer Churn Prediction & Retention Analysis
PythonPandasScikit-learnPower BIDAX
View on GitHub ↗
Analyzed telecom customer behavior to identify churn patterns, high-risk customers, and factors contributing to customer churn using Python, machine learning, and Power BI.
7,032customers analyzed
26.58%churn rate
$456Kmonthly revenue at risk
My contribution
- — Cleaned and prepared the IBM Telco Customer Churn dataset using Python and Pandas.
- — Created DAX KPIs for total customers, churned customers, churn rate, monthly charges, and revenue lost due to churn.
- — Built an interactive Power BI dashboard analyzing churn by contract type, tenure, internet service, and payment method.
Healthcare Predictive Analytics
PythonPandasScikit-learnPower BIDAX
View on GitHub ↗
Developed a healthcare analytics dashboard using patient data to analyze heart disease risk factors, health trends, and high-risk patient groups.
918patients analyzed
55.3%disease percentage
5+health analyses
My contribution
- — Cleaned healthcare data, handled missing values, removed duplicates, and encoded categorical variables.
- — Created DAX measures for total patients, heart disease patients, disease percentage, average cholesterol, and average MaxHR.
- — Built interactive analysis for blood pressure, cholesterol, chest pain type, gender, and high-risk patients.
House Price Analytics Dashboard
PythonPandasScikit-learnLinear RegressionPower BI
View on GitHub ↗
Combined machine learning and Power BI to analyze housing market trends, predict house prices, and compare actual versus predicted values.
108houses analyzed
0.85R² score
ML + BIintegrated workflow
My contribution
- — Preprocessed housing data, handled null values, encoded categorical variables, and selected relevant features.
- — Trained a Linear Regression model and generated predicted house prices.
- — Built a Power BI dashboard with actual vs predicted prices, price distribution, feature analysis, and interactive filters.
Google Play Store Analytics Dashboard
PythonPandasPower BIDAXJupyter
View on GitHub ↗
Analyzed Google Play Store applications to uncover trends in categories, installs, ratings, pricing, reviews, and user sentiment through an interactive Power BI dashboard.
8,196apps analyzed
75B+total installs
2B+total reviews
My contribution
- — Cleaned and standardized app, install, pricing, rating, review, and sentiment data using Python and Pandas.
- — Created DAX measures for total apps, average rating, total installs, and total reviews.
- — Built interactive dashboards for app categories, free vs paid applications, ratings, installs, and sentiment analysis.
Developed a Python desktop application that enables non-technical users to analyze CSV and Excel datasets through automated reports, visualizations, and export features.
CSV + Exceldata sources
4chart types
EXEstandalone application
My contribution
- — Built a Tkinter GUI for loading and validating CSV and Excel datasets.
- — Implemented automatic column detection and interactive GroupBy report generation.
- — Added Bar, Column, Line, and Pie chart generation with Excel/CSV and PNG export functionality.
- — Packaged the application as a standalone Windows executable using PyInstaller.
