tropicalberto/expedia_ranking_hotels

A hotel recommender system based on a Machine Learning and Data Mining approach

Machine Learning recommender system

Repository for the Data Mining course (VU University, 2017)

Description

We designed and trained a recommender system based on a Machine Learning approach. The algorithms are able to predict what hotel a user is most likely to book. The whole implementation is based in Python2, using data science packages (scikit-learn, numpy, pandas, ...).

The trained algorithms include:

• Random Forest regression (RF)
• Support Vector Machine (SVM)
• LambdaMART, based on Gradient Boosted Regression trees
• Coordinate Ascent

The overall implementation is exhaustively detailed in the report included in the report directory: https://github.com/tropicalberto/expedia_ranking_hotels/blob/master/report/report__hotel_recommender_system_v1.pdf

Dataset

A public dataset, released by Expedia, was retrieved from Kaggle (https://www.kaggle.com/c/expedia-personalized-sort)

Authors:

• Marie Corradi @MarieCo (https://github.com/MarieCo)
• Elena Garcia @egarcialara (https://github.com/egarcialara)
• Alberto Gil @tropicalberto (https://github.com/tropicalberto)

All authors contributed equally to this work.