A Cognitive Diagnosis Approach for Recommending Items Based on Polytomous Responses and Latent Attributes

• Autor(a) principal: Marana, Fernanda Tostes
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: https://www.teses.usp.br/teses/disponiveis/55/55134/tde-28082023-131450/
• topic: Atributos latentes; Cognitive diagnostic model; Hybrid model; Latent attributes; Modelo de diagnóstico cognitivo; Modelo híbrido; Polytomous response; Recommendation system; Respostas politômicas; Sistemas de recomendação

Recommendation Systems have become prevalent in recent years, attracting the attention of researchers to investigate different methods to filter relevant information for users. This information is not always explicit and different proposals have emerged to obtain the latent values of individuals through their behavior. In educational areas, latent attributes of test-takers can be acquired by psychometric models such as the Cognitive Diagnostic Model. These models attempt to create a users profile in order to explore the connections between students and subjects, just like a recommendation system does with its users and the products to be recommended. The objective of this work is to develop a new recommendation approach that incorporates Cognitive Diagnostic Models applied to data from media defined by discrete content (such as genres in movies and series) in order to generate its polytomous response in the form of the rating prediction that a user would give to each item. The proposed approach was applied to two datasets (MovieLens 20M Dataset and Anime Recommendation Database) and, due to the sparsity of the data, obtained in some cases better results than a classic content-based filtering recommendation method. Then, our new recommendation approach was fused to the classic recommendation model and this hybrid recommendation system obtained some results that were better when compared with the ones acquired by the individual systems. Finally, this work also explored the performance of the models in ranking items to be recommended for the users. Some interesting points were observed and the proposed model had the best performance even compared to the hybrid model.