Generating Stereotypes Automatically For Complex Categorical Features

TL;DR

This paper proposes a new method for automatically generating stereotypes for complex categorical features in recommender systems, improving clustering performance and addressing cold-start problems without relying on rating data.

Contribution

It introduces an algorithm that creates stereotypes based on feature metadata similarities, outperforming traditional clustering methods like k-modes for complex categorical data.

Findings

The proposed method outperforms k-modes in clustering complex categorical features.

Stereotypes generated can help mitigate cold-start issues.

Results validated on MovieLens and IMDb datasets.

Abstract

In the context of stereotypes creation for recommender systems, we found that certain types of categorical variables pose particular challenges if simple clustering procedures were employed with the objective to create stereotypes. A categorical variable is defined to be complex when it cannot be easily translated into a numerical variable, when the semantic of the categories potentially plays an important role in the optimal determination of stereotypes, and when it is also multi-choice (e.g., each item can be labelled with one or more categories that may be applicable, in a non pre-defined number). The main objective of this paper is to analyse the possibility of obtaining a viable recommendation system that operates on stereotypes generated directly via the feature's metadata similarities, without using ratings information at the time the generation of the classes. The encouraging results using integrated MovieLens and Imdb data set show that the proposed algorithm performs better than other categorical clustering algorithms like k-modes when clustering complex categorical features. Notably, the representation of complex categorical features can help to alleviate cold-start issues in recommender systems.