Sumit SC Portfolio: Projects, Tools, and Techniques

An overview of the project categories, analytical techniques, and Python tools used across this data science portfolio, from regression models to time-series.

This portfolio brings together data science and machine learning work spanning predictive modeling, exploratory data analysis, time-series analysis, and data visualization. Whether you are exploring a specific project or evaluating the breadth of techniques applied, you will find detailed write-ups, notebooks, and findings for each case study. The projects draw from real-world domains — automotive pricing, financial risk, entertainment analytics, and environmental detection — giving each analysis practical relevance beyond academic exercises.

Project categories

The portfolio is organized around four core types of analytical work:

• Predictive modeling — building and evaluating regression and classification models to generate actionable predictions
• Exploratory data analysis (EDA) — uncovering patterns, distributions, and relationships within raw datasets
• Time-series analysis — examining trends, seasonality, and volume patterns over time
• Data visualization — communicating findings clearly through charts, plots, and summary statistics

Featured projects

Used Car Price Prediction

ML regression model to estimate used car market prices based on vehicle attributes and historical data.

IMDB Movie Analysis

Exploratory analysis of movie ratings, genres, and revenue trends across decades of IMDB data.

Bank Loan Case Study

Risk analysis identifying key factors behind loan defaults to support data-driven credit decisions.

ABC Call Volume Trend

Time-series analysis of inbound call patterns to identify peak periods and operational trends.

Additional projects — including Stock Analysis and Oil Spill Detection — are also available in the portfolio.

Tools and technologies

Every project in this portfolio is built on a standard Python data science stack:

Tool	Purpose
Python	Core programming language for all data work
Pandas	Data loading, cleaning, and transformation
NumPy	Numerical computing and array operations
Scikit-learn	Machine learning model training and evaluation
Matplotlib	Base plotting and figure generation
Seaborn	Statistical visualization built on Matplotlib
Jupyter Notebooks	Interactive development and reproducible analysis

About this site

This portfolio is hosted on GitHub Pages via the Sumit-SC repository. Pages are built from Jupyter notebooks and markdown write-ups committed to the source repository. Cloudflare handles CDN delivery, and reader comments are powered by Giscus. See the deployment guide for the full hosting setup.