I Spanish Conference on Recommender Systems - SISREC 2024, Corunna (Spain). 19-21 June 2024
Summary:
This article is a summary of the work published in the journal Data Mining and Knowledge Discovery[1]. It presents an analysis of different types of biases and polarization measurements that affect the area of Point-Of-Interest recommendation. Our results evidence which state-of-the-art recommender ssuffer from popularity bias,venue exposure polarization,and geograph-ical distance polarization.
Spanish layman's summary:
Este artículo presenta un análisis de diferentes sesgos y medidas de polarización que afectan al ámbito de la recomendación de puntos de interés. Nuestros resultados evidencian qué recomendadores del estado del arte sufren de sesgo de popularidad, polarización de la exposición y polarización de la distancia geográfica.
English layman's summary:
This paper presents an analysis of different biases and polarization measurements that affect the area of Point-Of-Interest recommendation.
Our results evidence which state-of-the-art recommenders suffer from popularity bias, venue exposure polarization, and geographical distance polarization.
Keywords: POI recommendation, Bias mitigation, Polarization, Temporal Evaluation
Published in CAEPIA 2024, pp: 615-616, ISBN: 978-84-09-62724-0
Publication date: 2024-06-21.
Citation:
P. Sánchez, A. Bellogín, L. Boratto, Measuring and mitigating biases in location-based recommender systems, I Spanish Conference on Recommender Systems - SISREC 2024, Corunna (Spain). 19-21 June 2024. In: CAEPIA 2024: XX Conferencia de la Asociación Española para la Inteligencia Artificial. A Coruña, 19 al 21 de Junio de 2024, ISBN: 978-84-09-62724-0
