Downloading: A Survey of Mining Online Public Opinions from Twitter on Political Domain Using Machine Learning Techniques
International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
www.ijsr.net | Open Access | Fully Refereed | Peer Reviewed International Journal

ISSN: 2319-7064



To prevent Server Overload, Your Article PDF will be Downloaded in Next Seconds

A Survey of Mining Online Public Opinions from Twitter on Political Domain Using Machine Learning Techniques

Amruta U. Tarlekar, Manohar K.Kodmelwar

Abstract: Informal conversation of public on social media (e.g. twitter, Facebook, replies to particular news) shed light into their experiences (opinions, feelings, and concerns) about the political parties/leaders. Such unstructured data can provide valuable knowledge to political parties and even to public that what is current scenario of politics within that region. Analyzing such data, however, can be challenging. The complexity of publics experiences about politics/political leaders reflected from social media content requires human interpretation. However, the growing scale of data demands automatic data analysis techniques. In this project, we are going to develop a workflow to integrate both qualitative analysis and large-scale data mining techniques. We focused on publics Twitter posts, different micro blogging websites where public post their reviews/opinions about political parties/leaders to understand issues and problems that they have with them (political parties/leaders). We are going to conduct a qualitative analysis on samples taken from tweets/micro blogs related to political parties/leaders to identify different sentiments that is negative as well as positive aspects of public. Based on these results, we are going to implement a multi-label classification algorithm to classify tweets/micro blogs. Reflecting publics reviews about particular political party/leader and we are going to use this algorithm to train detector which will automatically detect sentiments (happy, sad, disgusting, and angry) from tweets and micro blogs.

Keywords: Information filtering, machine learning, Positive/Negative aspects of data, Sentiment detection



Top