Voice based Public Opinion Mining and Automatic Category System
Arunkanth. C. P · International Journal for Research in Applied Science and Engineering Technology · 2018
Nowadays, central and state governments allocates/spends heavy budget for people beneficial schemes.Even though there is no maintenance been carried and the schemes are not beneficial to the public hence people face several problems in day to day life.The People being unaware about the norms and procedure to approach the respective government departments that is responsible for the specific issue.Because of the above problem public and government negligence several lives are taken off. To overcome the considerable serious problems, voice based public opinion mining application is developed. In the proposed system the public can give their opinions in the web application with voice recognition module using their smartphones. For speech to text conversion, a Hidden Markov Models (HMMs) are used and the text is fed to the sentimental analysis module, in which preprocessing, part of speech, unigram analysis and fuzzy logic (machine learning technique) is used to forward the user opinion/issue to the concerned government department. The system provides two methods, Automatic e-mail alert to the government officials and a Report generation, where public and both government officials can view the issue which are opened and closed. The system tries to provide problems faced by the public in region / area wise, so that with the aggregated data the government would know which areas / region needs more attention. Thus the proposed system created a close relationship and communication mode with the government representatives elected by people of India.