Recommendation

Youtube search... ...Google search

• Capabilities
• Clustering
• AI Solver
• Collaborative Filtering | Wikipedia
• Personalized Marketing | Wikipedia

Can items can be directly related to users?

• ... yes, K-Nearest Neighbors (KNN)
• ... no, have a very large dataset, then Alternating Least Squares (ALS)
• ... no, have small to medium dataset, then Matrix Factorization

1. Content-based

systems examine attributes of the items recommended. For example, if a Netflix user has watched many cowboy movies, then recommend a movie classified in the datastore as having the “scifi” genre.

2. Collaborative Filtering (CF)

systems recommend items based on similarity measures between users and/or items. The items recommended to a user are those preferred by similar users.

Recommender Systems

• Recommender System | Wikipedia
• Recommendation System Algorithms | Daniil Korbut
• Business Intelligence Through Intellectual Property Analytics – Examining Facebook and Amazon | Valuenex
• Recommendation Systems | Stanford InfoLab
• Introduction To Recommendation system In Javascript | Oni Stephen - Medium

Grouping related items together without labeling them, e.g. grouping patient records with similar symptoms without knowing their symptoms

Building Recommendation Systems in Azure

• Microsoft Azure