I'm a Data Analyst working with SQL, Python, Excel, and Power BI to turn raw data into business insights. My focus areas are transaction analytics, customer segmentation, and KPI dashboard development — basically anything that involves finding patterns in messy data and explaining what they mean for a business.
Most of my project work is here on GitHub: UPI transaction and fraud analysis, customer RFM segmentation, retail SQL reporting, Power BI dashboards, and a few machine learning applications I built to understand how a model goes from raw data to a working prediction.
I like projects that end in a working deliverable — a dashboard, a query, or a small web app — rather than stopping at analysis on paper.
Data Analysis: SQL · MySQL · Python · Advanced Excel
Visualization & BI: Power BI · Tableau · Power Query · DAX · Plotly
Python Libraries: Pandas · NumPy · SciPy · Matplotlib · Scikit-learn
Analytics: Exploratory Data Analysis · Statistical Analysis · Data Cleaning & Validation · Dashboard Development · KPI Reporting · RFM / Customer Segmentation · Business Intelligence
Tools: Git · GitHub · Jupyter Notebook · VS Code
End-to-end analysis of 100,000 UPI transactions to uncover fraud risk and transaction-performance patterns.
Tech stack: SQL · Python · Excel · Power BI · SciPy Key insight: Rooted devices showed a fraud rate nearly 15x higher than non-rooted devices (20.69% vs. 1.39%), confirmed through statistical hypothesis testing.
A web app for analyzing Indian and global stocks using Yahoo Finance data, with moving averages, historical trends, and interactive charts.
Tech stack: Python · yfinance · Pandas · Plotly
🔗 Repository · Live Demo
Statistical analysis of retail customer data to understand demographics, spending behavior, and engagement, with recommendations for retention and targeted marketing.
Tech stack: Python · Pandas · NumPy · Matplotlib · Statistics
A Logistic Regression model that predicts whether a loan applicant is likely to repay or default, deployed as a web app with batch CSV predictions.
Tech stack: Python · Scikit-learn · Pandas · Logistic Regression
🔗 Repository · Live Demo
Segmented customers into groups such as Champions, Loyal, At-Risk, and Lost using RFM (Recency, Frequency, Monetary) analysis, with marketing recommendations for each segment.
Tech stack: Python · Pandas · NumPy · Matplotlib
A MySQL project managing customer, product, order, and payment data — covering joins, subqueries, aggregations, and business reporting.
Tech stack: MySQL · SQL
More projects
| Project | Description | Tech |
|---|---|---|
| Amazon Sales Analysis Dashboard | Excel-based sales analysis with pivot tables, pivot charts, slicers, and KPI tracking | Excel |
| Airline Performance Dashboard | Power BI dashboard analyzing airline revenue, delays, passenger load, and route profitability | Power BI, Power Query, DAX |
| Social Media Addiction & Academic Impact Dashboard | Power BI dashboard exploring the link between student social media usage and academic performance | Power BI, Power Query, DAX |
| Hybrid Vehicle Purchase Prediction | Logistic Regression web app predicting hybrid vehicle purchase intent from customer data | Python, Scikit-learn |
| AI-Based Emotion Detection | Flask learning project that detects emotion in text via IBM Watson NLP, with unit tests and a 10/10 Pylint score | Python, Flask, NLP |
| Sentiment Analysis | Classifies customer reviews as positive, negative, or neutral, and visualizes common complaint terms with a word cloud | Python, TextBlob, WordCloud |
- IBM Data Analytics Professional Certificate
- SQL for Data Science
- Generative AI for Business Intelligence
- Career247 Data Analyst with GenAI
- Python & Flask Certification on AI Application
📫 akshatraghav727@gmail.com 💼 linkedin.com/in/akshat-raghav727
Open to Data Analyst / Business Analyst opportunities.