cf2vec: Collaborative Filtering algorithm selection using graph distributed representations

TL;DR

This paper introduces cf2vec, a graph-based representation learning approach for Collaborative Filtering algorithm selection, which outperforms traditional metafeatures by being less data-intensive and more automated.

Contribution

It proposes a novel graph2vec-based method for representing collaborative filtering problems, reducing reliance on human-crafted metafeatures and improving robustness.

Findings

Representations are competitive with state-of-the-art metafeatures

Method requires significantly less data

Approach minimizes human input

Abstract

Algorithm selection using Metalearning aims to find mappings between problem characteristics (i.e. metafeatures) with relative algorithm performance to predict the best algorithm(s) for new datasets. Therefore, it is of the utmost importance that the metafeatures used are informative. In Collaborative Filtering, recent research has created an extensive collection of such metafeatures. However, since these are created based on the practitioner's understanding of the problem, they may not capture the most relevant aspects necessary to properly characterize the problem. We propose to overcome this problem by taking advantage of Representation Learning, which is able to create an alternative problem characterizations by having the data guide the design of the representation instead of the practitioner's opinion. Our hypothesis states that such alternative representations can be used to replace standard metafeatures, hence hence leading to a more robust approach to Metalearning. We propose a novel procedure specially designed for Collaborative Filtering algorithm selection. The procedure models Collaborative Filtering as graphs and extracts distributed representations using graph2vec. Experimental results show that the proposed procedure creates representations that are competitive with state-of-the-art metafeatures, while requiring significantly less data and without virtually any human input.