CaseRecommender 0.0.14

A recommender systems framework for Python

Case Recommender: A Recommender Framework for Python

Case Recommender is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback. The framework aims to provide a rich set of components from which you can construct a customized recommender system from a set of algorithms. Case Recommender has different types of item recommendation and rating prediction approaches, and different metrics validation and evaluation.

Algorithms

Item Recommendation:

• BPR MF

• Item KNN

• Item Attribute KNN

• User KNN

• User Attribute KNN

• Ensemble BPR Learning

• Most Popular

• Random

Rating Prediction

• Matrix Factorization (SVD)

• Item KNN

• Item Attribute KNN

• User KNN

• User Attribute KNN

• Item NSVD1 (with and without Batch Mode)

• User NSVD1 (with and without Batch Mode)

Evaluation and Validation Metrics

• All-but-one Protocol

• Cross-fold- Validation

• Item Recommendation: Precision, Recall and Map

• Rating Prediction: MAE and RMSE

• Statistical Analysis (T-test and Wilcoxon)

Requirements

• scipy

• numpy

More Information

https://github.com/ArthurFortes/CaseRecommender