Sentimental Analysis of Twitter Health Data using Machine Learning Techniques
E.M.Roopa Devi1, R.Rajadevi2, S.Vinoth Kumar3
1E.M.Roopa Devi*, Information Technology, Kongu Engineering College, Perundurai, India.
2R.Rajadevi, Information Technology, Kongu Engineering College, Perundurai, India.
3S.Vinoth Kumar, Information Technology, Kongu Engineering College, Perundurai, India.
Manuscript received on November 19, 2019. | Revised Manuscript received on 20 November, 2019. | Manuscript published on December 10, 2019. | PP: 3458-3462 | Volume-9 Issue-2, December 2019. | Retrieval Number: B6460129219/2019©BEIESP | DOI: 10.35940/ijitee.B6460.129219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: With the huge development of Internet, more users have occupied with wellbeing networks, for example, medicinal discussions to assemble wellbeing related data, to share encounters about medications, treatments, analysis or to associate with different clients with comparable condition in social media. A lot of lookup has focused on examining Twitter health tweets for subject matter modeling using quite a number clustering approaches, but few have mentioned it for sentiment analysis. The truth that such statistics carries potential information for revealing the opinion of humans about fitness services and behaviors make it an interesting study. In these paper, universal sentiments about Twitter health data was investigated. Twitter, measuring and monitoring the occurrence of social health problems. The approach is based on two stages: In first stage separating perhaps applicable tweets utilizing a lot of uniquely made standard articulations, and afterward arranges these underlying messages utilizing machine learning techniques. Using the Twitter search API and Twitter metadata geocoded content, social media tweets were selected to start filtering. Once Tweets are correctly identified, the classifier was applied to data in order to filter out the tweets. Classification results were improved by detecting the values of ROC and f-measure. This report indicates that such a method provides a viable solution for quantifying and tracking the progression of health status within society.
Keywords: Twitter Health Information, Sentimental Analysis, Feature Selection, Machine learning Techniques.
Scope of the Article: Machine learning