Vulcont: A Recommender System based on Contexts History Ontology
The use of recommender systems is already widespread. Everyday people are exposed to different items’ offering that infer their interest and anticipate decisions. The context information (such as location, goals, and entities around a context) plays a key role in the recommendation’s accuracy. Exten...
| Autor: | |
|---|---|
| Tipo de recurso: | tesis de maestría |
| Estado: | Versión publicada |
| Fecha de publicación: | 2017 |
| País: | Brasil |
| Institución: | Universidade do Vale do Rio dos Sinos (UNISINOS) |
| Repositorio: | Repositório Institucional da UNISINOS (RBDU Repositório Digital da Biblioteca da Unisinos) |
| Idioma: | portugués |
| OAI Identifier: | oai:www.repositorio.jesuita.org.br:UNISINOS/6352 |
| Acceso en línea: | http://www.repositorio.jesuita.org.br/handle/UNISINOS/6352 |
| Access Level: | acceso abierto |
| Palabra clave: | ACCNPQ::Ciências Exatas e da Terra::Ciência da Computação Sistemas de recomendação Ontologia Histórico de contextos Filtragem colaborativa Recommender systems Ontology Contexts history Collaborative filtering |
| Sumario: | The use of recommender systems is already widespread. Everyday people are exposed to different items’ offering that infer their interest and anticipate decisions. The context information (such as location, goals, and entities around a context) plays a key role in the recommendation’s accuracy. Extending contexts snapshots into contexts histories enables that information to be exploit. It is possible to identify context’s sequences, similar contexts histories and even predict future contexts. In this work we present Vulcont, a recommender system based on a contexts history ontology. Vulcont merges the benefits of ontology reasoning with contexts histories in order to measure contexts history similarity, based on semantic and ontology’s properties provided by context’s domain. Vulcont considers synonymous and classes’ relations to measure similarity. After that, a collaborative filtering approach identifies sequences’ frequency to identify potential items for recommendation. We evaluated and discussed the Vulcont’s recommendation in four scenarios in an offline experiment, which presents Vulcont’s recommendation power, due the exploit of semantic value of contexts history. |
|---|