Instrumento para mensurar privacidade em ambientes IoT

• Autor(a) principal: Lopes, Bruno
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/12159
• topic: Internet das coisas; Privacidade; Preocupações de privacidade; Análise fatorial; Internet of things; Privacy; Privacy concerns; Factorial analysis; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO

The Internet of Things (IoT) connects devices that are commonly used in people's daily lives - such as cell phones, televisions, coffee makers, refrigerators, beds, sensors, and more - so that they automatically communicate over a network. Generic, impersonal information exchange between devices can occasionally lead to privacy issues, such as making personal information available to applications when using them. Since it retains a concept that involves various dimensions of data considered private, such as body, behavioral, communication and personal. To measure privacy in IoT environments, this paper presents the design of an instrument, called the Internet of Things Privacy Concerns (IoTPC), which is capable of reflecting users' concerns about privacy in an IoT environment. The IoTPC instrument consists of 17 items obtained through an analysis of the privacy measurement instruments available in the current literature, which include users' opinion on how devices collect, process and make their personal information available in specific IoT scenarios. . The validation of the IoTPC was performed from the analysis of the results of a sample of 61 participants, considering the dimensions of IoT requests, decision power and caution, through exploratory factor analysis. IoTPC subsidized the construction of an inference module in a privacy negotiation mechanism for IoT systems. This module performs an inference based on IoTPC items and IoT scenarios through machine learning algorithms, which have been trained and tested with the privacy preferences derived from the IoTPC instrument. The results of the learning process of the inference module obtained an accuracy of 79.20%, concluding that the instrument can be employed by a privacy negotiation mechanism.