MADS: Model Analysis & Decision Support

📖Notes and remarks on Machine Learning related papers

Recommendations with IBM Data (knowledge-based plus collaborative filtering both model-based and neighborhood-based)

Performed EDA, created user-article matrix, calculated similarity using dot product, implemented Rank-Based, User-User CF, Content-Based, and Matrix Factorization, evaluated model with precision, recall, and F1-score.

In the IBM Watson Studio, there is a large collaborative community ecosystem of articles, datasets, notebooks, and other A.I. and ML. assets. Users of the system interact with all of this. This is a recommendation system project to enhance the user experience and connect them with assets. This personalizes the experience for each user.

Numerical Analysis Projects

Project 3 in Data Scientist Nanodegree with Udacity. Build a recommender engine for IBM Watson.

Recommendation System for IBM articles

EDA, Pre-processing, 6 Recommendation Systems Techniques: * Popularity-Based, * Cosine Similarity Collaborative Filtering, * Matrix Factorization Collaborative Filtering, * Clustering, * Content-Based Filtering, * Hybrid Recommendation System.

Evaluating k-nearest neighbors and singular value decomposition techniques for collaborative filtering recommender systems

A Text / Speech Summarizer

Predicting Nobel Physics Prize winners. Final project for Harvard CS109a 2017 edition https://github.com/covuworie/a-2017.

analyze the interactions that users have with articles on the IBM Watson Studio platform, and make recommendations on new articles they will like.

Complete concepts behind implementing a Recommendation System using Association Rules, Collaborative Filtering, and Matrix Factorization.

Analyze the interactions that users have with articles on the IBM Watson Studio platform, and make recommendations to them about new articles.

Recommender system from Yelp dataset

A Cloud Based Personalised Recommendation System for movies and books.

Recommendation engine for IBM Watson.

This project is to analyze the interactions that users have with articles on the IBM Watson Studio platform, and make recommendations to them about new articles they might like. Recommending articles that are most pertinent to specific users is beneficial to both service providers and users.