Identificando Emoções em Textos em Português do Brasil usando Máquina de Vetores de Suporte em Solução Multiclasse

Suportando Contratos de QoS no Nível Arquitetural

Mariza Miola DosciattiLohann Paterno Coutinho FerreiraEmerson Cabrera Paraiso

S. AnsaloniA. SztajnbergR. CerqueiraO. G. Loques

The automatic identification of emotions in texts has shown significant results in several applications. In this article, we present an approach using Support Vector Machines to identify emotions in texts written in Brazilian Portuguese. The corpus used in the experimentconsists of news extracted from an online newspaper. The texts were labeled and subjected to a SVM classifier in a multiclass configuration, obtaining an accuracy rate of 61%. This paper presents an approach to describe, deploy and manage component-based applications having dynamic functional and non-functional requirements, which include different types of QoS. The approach is centered on architectural descriptions and associated high-level QoS contracts. These contracts are used to guide configuration adaptations required to enforce QoS requirements. The infrastructure required to manage the contracts follows an architectural pattern, which can be directly mapped to specific components included in a supporting reflective middleware. This allows designers to write a contract and to follow standard recipes to insert the extra code required to its enforcement in the supporting middleware.

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato:
     Mantida por: