Social Media Intelligence extractor

Abstract

In the last few centuries, information has highly appreciated in value and its sources have greatly increased in number. The introduction of the web 2.0, which allowed users not only to retrieve but also generate information as authors, gave birth to a whole new data source called social media. Social media is a term that integrates technology, social interaction and user generated content and differs from traditional broadcasting. It commonly comprises of technologies such as instant messaging programs, discussion forums, weblogs and wikis. Facebook, YouTube and Wikipedia are examples of popular social media websites. The information generated is greatly useful to individuals, organizations, institutions and governments globally. But this information tends to be highly unstructured and certain times not trustworthy. Social Computing is a novel and emerging computing paradigm that involves a multi-disciplinary approach in analyzing and modeling social behaviors on different media and platforms to produce intelligent and interactive applications and results. The objective of this project is to summarize social media opinions on various subjects with the focus on Twitter and Facebook microblog systems. In this project, we propose and attempt to implement a system that uses various machine learning and computational techniques used in Social Computing to collect, extract, process, mine, and visualize the data. This summarization task is different from traditional text summarization because we are only interested in the positive, negative and neutral opinions people have expressed on specific features or topics. This will be done at both the sentence and at the post level (document level).Keywords: Social media, web crawling, text processing, data mining, machine learning, topic detection, sentiment analysis, quality content detection.