Social Data Mining: Collection and Analysis of Political Posts
Ahmed Chaari · ERA: Education and Research Archive (University of Alberta) · 2016
Social Networks become an important part of people's lives. A large number of individuals contribute to their contents. All aspects of everyday life -- from work to politics, and further from entertainment to personal events -- are reflected in posts generated on a daily basis. Those posts are perceived as a source of information that becomes a target and subject of analysis and research. This thesis addresses an issue of collecting and analyzing tweet posts. A comprehensive study of available programming tools suitable for collecting posts from multiple social networks has been conducted. Based on the results of that investigation, we have designed and developed a methodology for collecting tweets. Further, we have constructed a Social Data Mining Platform. The platform, implemented using Elasticsearch (open source full-text search engine), provides a number of approaches and algorithms for analysis of tweets. The analysis goes beyond processing of hast-tags and includes matching processes that involve the whole context of tweets. We have selected political tweets as the subject of our study. We have demonstrated how the proposed platform can be used to gain better understanding of political issues and opinions of a populace.