Python Data Mining & Applied Algorithms

This repository contains Python implementations of advanced data mining algorithms, machine learning pipelines, and high-performance recommender systems.

Core Implementations & Performance

• Recommender Systems (02_recommender_systems_svd.ipynb): Engineered scalable User Collaborative Filtering and Singular Value Decomposition (SVD) models.
• Performance Optimization: Achieved sub-5-second execution times on 100k+ row datasets by strictly utilizing sparse matrices (scipy.sparse.csr_matrix) and vectorized mathematical operations via NumPy, successfully eliminating iterative Python loops.
• Machine Learning & NLP (03_clustering_nlp_pipelines.ipynb): Implemented classification pipelines utilizing TF-IDF and pre-trained Word2Vec models for NLP feature extraction. Designed clustering models (K-means, Agglomerative) using Silhouette coefficients.
• Data Engineering (01_data_preprocessing_eda.ipynb): Cleaned and transformed raw, high-dimensional data using Pandas for downstream machine learning tasks.

Technology Stack

• Language: Python
• Libraries: Pandas, NumPy, SciPy, Scikit-learn, Matplotlib