CS2 Match and Demo Analytics
A data pipeline and analysis tool for CS2 esports match and player performance using HLTV data and demo file parsing. The system scrapes matches, processes player stats, and supports ML-driven performance insights.
Tags: Python, PostgreSQL, Web Scraping, Streamlit, Data Engineering, ETL
Support Ticket Categorization with LLM
Used a Large Language Model to classify and prioritize customer support tickets. The system predicts categories, assigns priority, and estimates resolution times.
Tags: Python, HuggingFace, Transformers, Mistral 7B, NLP, Text Classification, LLMs
Bank Churn Prediction
Predicted bank churn using ANNs. Achieved 78% recall score to minimize false negatives, resulting in only 5.75% false negatives on the test data.
Tags: Python, TensorFlow, Keras, ML Classification, ANN, Feature Engineering
Credit Card User Churn Prediction
Predicted CC users churn by using multiple classifiers and tuning the best performing. Determined most important factors contributing to CC users churn. Recall was maximized to minimize the number of CC users that would churn, resulting in 0.99% false negatives.
Tags: Python, Scikit-Learn, ML Classification, Data Cleaning, Precision Optimization