POMDP planning for the creation of service packages

The generation of complex systems where the user requires decisions based on several criteria involves building a relationship of global preference for a set of alternatives evaluated. Between these systems there are the Tourist Systems, that was impulse by internet like a facilitator in the access...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Ferrarini Oliver, Cintia, Aciar, Silvana, Forradellas Martinez, Raymundo Quilez
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2014
Materias:
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/41853
http://43jaiio.sadio.org.ar/proceedings/IJCAI/29-30.pdf
Aporte de:
Descripción
Sumario:The generation of complex systems where the user requires decisions based on several criteria involves building a relationship of global preference for a set of alternatives evaluated. Between these systems there are the Tourist Systems, that was impulse by internet like a facilitator in the access of different sources of information for travel planning. Currently, tourists should not only visit multiple independent Web sites (airlines, hotels, etc.) and await a considerable time receiving confirmation responses. All these actions required generating a stochastic system that can meet the services and generate plans. Are being studied POMDP algorithms, in order to obtain a method for packages tailored to user requirements. Under the hypothetical-deductive methodology, this includes the study of service pakage´s composition, application, comparison and validation by a computer prototype with real users for a Web Recommender System. This thesis is framed within a PICT project called "E-OASIS Integrated Travel Services to facilitate easy access to Local Cultural Heritage" whose technical objective is to develop a Customized Recommender System for Tourist Services in the Province of San Juan. The automatic planning in real world problems has become a subject of interest to the scientific community because it allows computationally establish courses of action in cases where a human cannot addressed adequately for the complexity. The motivation for this work focuses on two aspects of the automatic planning, the scientific contributions and its practice application. <i>(Párrafo extraído del texto a modo de resumen)</i>