Publicação
Grouplanner: a group recommender system for tourism with multi-agent MicroServices
| Resumo: | To provide recommendations to groups of tourists is a very complex task, especially due to conflicting preferences and the group’s heterogeneity. The introduction of Multi-Agent Systems (MAS) can be the leverage we are looking for. Their autonomy, isolated state, distribution, and loose coupling make them suitable for the development of distributed systems, being the concept similar to a Microservices architecture. This connection brought a new approach, the Multi-Agent Microservices (MAMS) architecture, which exposes agents as resources through REST endpoints, changing the way MAS are seen and implemented, facilitating the user ↔ agent interaction, with a more efficient interoperability, bringing faster and more intelligent systems. In this demonstration, we propose the use of a MAMS architecture to represent the tourists in a mobile Group Recommender System for Tourism prototype, Grouplanner, exposing their agents and knowledge as resources that can be consumed by HTTP clients by directly communicating with the tourists’ agents using REST endpoints, in order to provide faster and better recommendations. |
|---|---|
| Autores principais: | Alves, Patrícia |
| Outros Autores: | Gomes, Domingos; Rodrigues, Catarina; Carneiro, João; Novais, Paulo; Marreiros, Goreti |
| Assunto: | Group recommender systems Multi-agent microservices Tourism |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | comunicação em conferência |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | To provide recommendations to groups of tourists is a very complex task, especially due to conflicting preferences and the group’s heterogeneity. The introduction of Multi-Agent Systems (MAS) can be the leverage we are looking for. Their autonomy, isolated state, distribution, and loose coupling make them suitable for the development of distributed systems, being the concept similar to a Microservices architecture. This connection brought a new approach, the Multi-Agent Microservices (MAMS) architecture, which exposes agents as resources through REST endpoints, changing the way MAS are seen and implemented, facilitating the user ↔ agent interaction, with a more efficient interoperability, bringing faster and more intelligent systems. In this demonstration, we propose the use of a MAMS architecture to represent the tourists in a mobile Group Recommender System for Tourism prototype, Grouplanner, exposing their agents and knowledge as resources that can be consumed by HTTP clients by directly communicating with the tourists’ agents using REST endpoints, in order to provide faster and better recommendations. |
|---|