Exploiting Lod-based Similarity Personalization Strategies for Recommender Systems

Enviado por hugo.chaves em seg, 28/08/2023 - 05:24

Nome completo do aluno

Gabriela Oliveira Mota da Silva

Título do trabalho

Resumo do trabalho

Linked Open Data (LOD) is a cloud of freely accessible and interconnected datasets that encompass machine-readable data. These data are available under open Semantic Web standards, such as Resource Description Framework (RDF) and SPARQL Protocol and RDF Query Language (SPARQL). One notable example of an LOD set is DBpedia, a crowd-sourced community effort to extract structured information from Wikipedia and make this information openly available on the Web. The semantic content of LOD and the advanced features of SPARQL have opened unprecedented opportunities for enabling semantic-aware applications including Recommender Systems. LOD-based Recommender Systems usually leverage the data available within LOD datasets such as DBpedia to recommend items such as movies, places, books, and music to end-users. These systems use a semantic similarity algorithm that calculates the degree of matching between pairs of resources by counting the number of direct and indirect links between them, the length of the path between them, or the hierarchy of classes. Conversely, calculating similarity in RDF graphs could be a difficult task because each resource can have hundreds of links to other nodes and not all of them are semantically relevant or can be applied to all resources in the graph. This can lead to the well-known matrix sparsity problem. Nevertheless, some effort has been made to select subsets of features, i.e., links, which are more helpful to computing similarity between items of a graph dataset, reducing the matrix dimension. Despite the existence of several studies in this field, there is still a lack of solutions applied to the personalization of feature selection tasks. In this context, we propose personalized strategies to improve semantic similarity precision on LOD-based Recommender Systems, including i) applying a feature selection approach to select the best features for a particular user; ii) personalization of the RDF graph by adding weights to the edges, according to the user’s previous preferences; and iii) exploiting the similarity of literal properties as well as the links from the user model. The evaluation experiments used combined data from DBpedia and MovieLens and from DBpedia and LastFM datasets. Results indicate significant increases in top-n recommendation tasks in Precision@K (K=5, 10), Map, and NDCG over non-personalized baseline similarity methods such as Linked Data Semantic Distance and Resource Similarity. The results show that the LOD-based strategies of user model personalization and feature selection explored in this work are efficient for improving content-based recommender systems in diverse contexts.

Orientador

Frederico Araújo Durão

Membro Titular Externo 1 (com afiliação)

Natasha Correia Queiroz Lino (UFPB)

Link para o curriculum lattes

http://lattes.cnpq.br/7853125713114677

Membro Titular Externo 2 (com afiliação)

Rosalvo Ferreira de Oliveira Neto (UNIVASF)

Link para o curriculum lattes

http://lattes.cnpq.br/9548186939653024

Membro Titular Interno 1 ou Titular Externo 3 (com afiliação)

Daniela Barreiro Claro (PGCOMP/UFBA)