Identificação de perfis de permissões em aplicativos móveis utilizando agrupamento e visualização

• Ano de defesa: 2019
• Autor(a) principal: Guiraldelli, Francisco Augusto Cesar de Camargo Bellaz
• Orientador(a): Faceli, Katti
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus Sorocaba
• Programa de Pós-Graduação: Programa de Pós-Graduação em Ciência da Computação - PPGCC-So
• Palavras-chave em Português: Perfil de Permissão; Aplicativos Móvies; Análise de Agrupamento; Ensembles de Agrupamento; Grupos Consenso; Visualização; Profile of Permissions; Mobile Applications; Clustering; Clustering Ensembles; Consensus Groups; Visualization
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/12237

In the last decade, mobile applications have gained widespread use through smartphones, it has radically transformed the way people can access and use the media, bringing the internet and many everyday tasks in the palm of their hands. Such ease, however, brought with it new challenges, such as the need for devices with high energy efficiency, good processing capacity, performance, good ergonomic form, lightweight and easy handling. Despite the great efforts expended by all of these companies, some issues with mobile application classification are at the top of many developers' concerns, as implementation defects, lack of domain knowledge or even preparation for future functionalities can cause mobile applications to consume hardware resources incorrectly allocated to both the application user's need and the domain for which it was developed. Focusing on the identification of permissions' profiles, the correct use of resources and the way the application is implemented, this research proposes a method that involves Google Play Store data collection, to use techniques like clustering and visualization, which will provide to device manufacturers, mobile application developers, and researchers a simple way to compare and analyze mobile applications that use similar features. To evaluate the method, a case study related to the energy consumption of mobile applications was conducted, which proves its effectiveness in this type of analysis.