Downloads: 124 | Views: 211
M.Tech / M.E / PhD Thesis | Computer Science & Engineering | India | Volume 8 Issue 4, April 2019
Intelligent System for Understanding and Monitoring of Health Diseases on Social Media
S. Kiruthika  | K. Suganthi
Abstract: Online networking has turned into a noteworthy hotspot for breaking down all parts of every day life. Because of devoted inactive theme investigation techniques. In this work, we are keen on utilizing internet based life to screen individuals' wellbeing after some time. The utilization of tweets has a few advantages including prompt information accessibility at practically no expense. Early observing of wellbeing information is integral to post-factum thinks about and empowers a scope of utilizations. We initially propose the Temporal Ailment Topic Aspect Model (TM? ATAM), another inert model committed to tackling the main issue by catching changes that include wellbeing related points. TM? ATAM is a non-clear augmentation to ATAM that was intended to extricate wellbeing related points. It learns wellbeing related theme changes by limiting the forecast blunder on subject disseminations between continuous posts at various time and geographic granularities. To take care of the second issue, we create T? ATAM, a Temporal Ailment Topic Aspect Model where time is treated as an irregular variable locally inside ATAM. Our examinations on a 8-month corpus of tweets demonstrate that TM? ATAM beats TM? LDA in assessing wellbeing related changes from tweets for various geographic populaces. We inspect the capacity of TM? ATAM to recognize changes because of atmosphere conditions in various geographic locales. At that point show how T? ATAM can be utilized to anticipate the most critical progress and also contrast T? ATAM and CDC (Center for Disease Control) information and Google Flu Trends.
Keywords: Social media, TM-ATAM, CDC, TM-LDA, Diseases
Edition: Volume 8 Issue 4, April 2019,
Pages: 1854 - 1858