"End-to-end Data Science and Machine Learning projects portfolio with Python, ML, Deep Learning, NLP, and deployment-ready solutions."
A comprehensive data analysis and machine learning project focused on predicting passenger survival on the Titanic
- Data Exploration & Cleaning
- Feature Engineering
- Model Building &Evaluation
- Visualizations & Insights
A complete end-to-end Machine Learning project to detect whether a news article is Real or Fake using NLP techniques, datasets, and deployment. A text classification project that applies Natural Language Processing (NLP) and Machine Learning to detect fake news from real news
- Data Exploration & Cleaning
- Text Preprocessing (tokenization, stopword removal, TF-IDF)
- Model Building & Evaluation
- Visualizations & Insights
- Deployment with Streamlit
The model is deployed using Streamlit and can be run locally or hosted on platforms like Streamlit Cloud.
A machine learning-powered web application that predicts house rental prices in Hyderabad, India.
Built with Streamlit and Random Forest Regression, this app provides accurate price estimates based on property characteristics.
- Data Cleaning & Preprocessing
- Feature Engineering
- Model Building & Evaluation
- Cross-Validation & Model Performance Metrics
- Interactive Web App with Streamlit
- Data Visualizations & Insights
A Natural Language Processing (NLP) and Machine Learning project focused on predicting whether two Quora questions are semantically similar (duplicates) or not.
The project leverages feature engineering, vectorization techniques, and multiple ML algorithms for evaluation.
- Data Exploration & Cleaning
- Advanced Feature Engineering (NLP features)
- Text Preprocessing (stemming, stopword removal, contractions expansion)
- Vectorization (TF-IDF, TF-IDF Word2Vec with GloVe)
- Model Training & Evaluation
- Visualizations (Word Clouds, t-SNE, PCA)
A Machine Learning classification project to predict whether a given sonar signal corresponds to a Rock or a Mine (Metal Cylinder).
This project leverages multiple ML models, feature visualization, and evaluation techniques.
- Data Exploration & Cleaning
- Class Distribution Visualization
- Feature Distribution Analysis
- Dimensionality Reduction (PCA)
- Model Building & Comparison
- Confusion Matrix, ROC, Precision-Recall Curves
The project is structured for local Jupyter Notebook execution, but can be extended for deployment using Streamlit or Flask if needed.
