A Text Analyser of Crowdsourced Online Sources for Knowledge Discovery

In the last few years, Twitter has become the centre of crowdsourced-generated content. Numerous tools exist to analyse its content to lead to knowledge discovery. However, most of them focus solely on the content and ignore user features. Selecting and analysing user features such as user activity and relationships lead to the discovery of authorities and user communities. Such a discovery can provide an additional perspective to crowdsourced data and increase understanding of the evolution of the trends for a given topic. This work addresses the problem by introducing a dedicated software tool developed, the Text Analyser of Crowdsourced Online Sources (TACOS). TACOS is a social relationship search tool that given a search term, analyses user features and discovers authorities and user communities for that term. For knowledge representation, it visualises the output in a graph, for increased readability. In order to show the applicability of TACOS, we have chosen a real example and aimed through two case studies to discover and analyse a specific type of user