Two-stage Cascaded Classifier for Purchase Prediction

TL;DR

This paper presents a two-stage cascaded classifier for purchase prediction in recommender systems, utilizing Random Forests and boosting to handle class imbalance and feature variability, achieving competitive results.

Contribution

The paper introduces a novel two-stage cascaded classifier approach that improves purchase session and item prediction efficiency and accuracy.

Findings

Random Forests effectively handle feature subset variability.

Boosting mitigates class imbalance in buy-session prediction.

Achieved competitive scores on the RecSys Challenge 2015 dataset.

Abstract

In this paper we describe our machine learning solution for the RecSys Challenge, 2015. We have proposed a time efficient two-stage cascaded classifier for the prediction of buy sessions and purchased items within such sessions. Based on the model, several interesting features found, and formation of our own test bed, we have achieved a reasonable score. Usage of Random Forests helps us to cope with the effect of the multiplicity of good models depending on varying subsets of features in the purchased items prediction and, in its turn, boosting is used as a suitable technique to overcome severe class imbalance of the buy-session prediction.