Desafio no uso e coleta de conhecimento cultural de aplicações para promover o crescimento da base de conhecimento cultural do OMCS-BR

• Ano de defesa: 2013
• Autor(a) principal: Bueno, André de Oliveira
• Orientador(a): Silva, Junia Coutinho Anacleto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC
• País: BR
• Palavras-chave em Português: Interação homem-máquina; Senso comum; Filtro cultural; Base de conhecimento; Contextualização; OMCS-Br; Cultura; Coletor de conhecimento
• Palavras-chave em Inglês: Culture; Common sense; Cultural filter; Knowledge collecto; Contextualization; Knowledgebase
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/547

We are currently in the third wave of research in the Human Computer Interaction (HCI) area, whose focus is on the context of applications. Technology has been spreading from the workplace to our homes; it becomes part of our everyday lives. Thinking about it, one way of trying to provide this contextualization can be done through culture insertion in applications. However, culture is a very broad concept that encompasses different factors, among them, the common sense, which is a subset of culture consisting of the knowledge we gain in our day-to-day routine activities. Within this context, this work aims to create a new strategy to culturally contextualize an application. Thinking about that, it was created a new module called Cultural Filter that, through the use of a given knowledgebase, filters the stored information to create small cultural contextualized slices according to the profile of the users who will use this application. In this case, information such as age, gender, education level, and geographic location are used in the process of creating these cultural slices. In order to observe the viability of this strategy, an instance of the cultural filter was instantiated using the cultural knowledgebase of project Open Mind Common Sense in Brazil (OMCS-Br). In this case, the filter acts creating cultural slices from the project complete knowledgebase. In the background, this also aims to further the collection of data for this database in question, collecting data generated from the use of these contextualized software. To make this set of cultural filter and knowledge collector accessible to developers and users in general, it was developed an online system that allows any interested person to use the knowledgebase of the project OMCSBr in its applications or, still, just do culturally contextualized searches in the base. We intend to expand access to the Brazilian cultural knowledgebase and make the collection and use of the OMCS-Br knowledgebase operations more flexible, facilitating the development of applications using such a database and, with the filter being part of the OMCS-Br architecture, applications become lighter, closer to the RIA (Rich Internet Application) concept. To achieve the proposed objectives, the OMCS-Br architecture was restructured and expanded. Two new modules have been developed in order to connect with the project API (Application Programming Interface), namely: (i) the Cultural Filter, responsible for performing the filtering in order to retrieve contextual data from the knowledgebase according to the predefined profile and (ii) the Knowledge Collector, responsible for entering collected data through the use of culturally contextualized software into the OMCS-Br project knowledgebase. As a proof of concept, the filter was used in some already existing applications from LIA and, therefore, it was possible to observe the appropriateness of its operation.