SWEETS: um sistema de recomendaÃÃo de especialistas aplicado a redes sociais

AUTOR(ES)

Edeilson Milhomem da Silva

DATA DE PUBLICAÇÃO

2009

RESUMO

The organizations aim at increasing industrial competitiveness. In this context, they have been searching for new ways to improve their productivity, the quality of their products, and costs reduction. To achieve these goals, it is essential to explore the collaboratorsâ potentials and the relationship among them, thus finding and sharing tacit knowledge. Since the tacit knowledge is embedded in the people s mind, it is therefore hard to be formalized and documented. This way identifying and recommending the person that retains the mentioned knowledge is the best option. In this context, this work presents the Specialist Recommendation System (SWEETS) and its application into the a.m.i.g.o.s environment, a knowledge management platform based on the social network concept. The SWEETS system has two versions, 1.0 and 2.0. The 1.0 version, puts people with specialties in common together proactively, either by their knowledge (according to the writing profile), or by their interest (according to the reading profile). On the other hand, the SWEETS second version does not act in a proactive way. Thereby, it is necessary to request for a specialist in a specific field, and it uses folksonomy to extract an ontology, which is essential to identify people s skills effectively. This ontology is reflected by the tags (concept) co-occurrence relating them to items (instances). In addition, such ontology is independent of domain, being the main contribution of this work. Applying the SWEETS system into the a.m.i.g.o.s. environment aims at bringing some benefits, such as: minimize the communication problem in the corporation, provide an encouragement of social knowledge and knowledge sharing; Therefore, a better usage of the collaborators knowledge may be expected.

ASSUNTO(S)

redes sociais web ciencia da computacao folsonomia e ontologia sistemas de recomendaÃÃo engenharia de software recommendation systems web-based social network, folksonomy ontology

Documentos Relacionados

• ICARE: um sistema de recomendaÃÃo de especialistas sensÃvel a contexto
• CoLORS: um sistema de recomendaÃÃo de oportunidades de aprendizado sensÃvel a contexto
• Uma ferramenta para recomendaÃÃo de clusters de mensagens para fÃbricas de software
• RecomendaÃÃo de comunidades virtuais em uma rede social para minimizar problemas de comunicaÃÃo entre projetos de software
• Uma soluÃÃo em filtragem de informaÃÃo para sistemas de recomendaÃÃo baseada em anÃlise de dados simbÃlicos