DataScience

View on GitHub

Data Science & Machine Learning Portfolio

This repository contains a comprehensive collection of Data Science, Machine Learning, Deep Learning, and Big Data Analytics projects implemented using Python, Jupyter Notebooks, Apache Spark, and TensorFlow.

The work spans classical ML algorithms, clustering, NLP, recommender systems, CNNs, streaming analytics, and real-world healthcare datasets, with an emphasis on hands-on modeling, evaluation, and interpretation.


🔧 Tools & Technologies


📂 Project Overview


1️⃣ Machine Learning Algorithms Portfolio Projects

📁 Machine Learning Algorithms Portfolio Projects/

Overview:
A comprehensive collection of hands-on machine learning projects implementing core supervised, unsupervised, and recommendation algorithms using Python and scikit-learn. Each notebook follows a complete workflow including data preprocessing, model training, evaluation, and interpretation on real-world datasets such as advertising data, loan data, Yelp reviews, MovieLens, and ecommerce data.

Key Focus Areas:


2️⃣ Linear and Logistic Regression, Clustering

📁 Linear and Logistic Regression,Clustering/

Overview:
This project focuses on the fundamental building blocks of machine learning, including linear regression, logistic regression, and clustering techniques. The notebooks emphasize understanding model assumptions, interpreting coefficients, and evaluating model performance using real datasets.

Key Focus Areas:


3️⃣ Convolutional Neural Networks (Dogs vs Cats)

📁 Convolutional Neural Networks/

Overview:
A deep learning project implementing Convolutional Neural Networks (CNNs) to classify images of dogs and cats. Multiple CNN architectures were trained and evaluated using TensorFlow, with saved models, checkpoints, and training logs included for experimentation and comparison.

Key Focus Areas:


4️⃣ Apache Spark — Streaming, MLlib & GraphFrames

📁 Apache Spark_Streaming, MLib, Graphframes/

Overview:
This project demonstrates big data analytics using Apache Spark, covering batch machine learning with Spark MLlib, real-time streaming analytics on NSE stock data, and graph analytics using GraphFrames. It highlights scalable data processing beyond single-machine workflows.

Key Focus Areas:


5️⃣ Web Scraping and Exploratory Data Analysis (EDA)

📁 Web Scrapping and EDA/

Overview:
A data acquisition and analysis project combining web scraping, data cleaning, and exploratory data analysis. The notebooks demonstrate how raw scraped data is transformed into structured datasets and analyzed using visualization and statistical summaries.

Key Focus Areas:


6️⃣ Wisconsin Breast Cancer Project

📁 Wisconsin Breast Cancer Project/

Overview:
A healthcare-focused data science project applying machine learning techniques to the Wisconsin Breast Cancer datasets. The project includes tumor classification, cancer recurrence prediction, and cancer survival analysis, supported by ROC curves, model comparisons, and detailed documentation.

Key Focus Areas:


📈 What This Portfolio Demonstrates

✔ Strong ML fundamentals
✔ Hands-on modeling and evaluation
✔ Deep learning with CNNs
✔ Spark-based big data analytics
✔ Healthcare & real-world datasets
✔ End-to-end DS workflows


📫 Contact


📄 License

This repository is licensed under the MIT License (see LICENSE).