TweeProfiles2: real time detection of spatio-temporal patterns in Twitter

• Autor(a) principal: João Henrique Alves Pereira
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: https://hdl.handle.net/10216/75800
• topic: Engenharia electrotécnica, electrónica e informática; Electrical engineering, Electronic engineering, Information engineering

With the advent of social networking a lot of user-specific, voluntarily provided data has been generated. A few years ago, people and companies started noticing the value that lied within those enormous amounts of information and started to think of ways to extract patterns from those and use them for gain. TweeProfiles is a visualization tool that allows analysing tweets' data over 4 dimensions: spatial, temporal, social and content. It is, however limited in the sense that it is not prepared to deal with streaming data. The goal of this work is to find, in real-time, patterns of information in data extracted from Twitter. The data to be harvested is multi-dimensional in nature, having namely a spatial dimension (the location of the tweet), a temporal dimension (the timestamp of the tweet) and a content dimension (the text and hashtags of the tweet). To achieve this goal, a clustering technique that is suitable for streaming data will be applied to the collected data; the resulting clusters will then be presented to the end-user through a visualization tool that is also scheduled to be developed in the scope of this project.