Turkish Journal of Computer and Mathematics Education
Journal license

Journal

Turkish Journal of Computer and Mathematics Education


Volume
& Issue

Volume 10, Issue 3


Published
on


Pages

1286-1290


DOI

Article

A Novel Hybrid Model for Stress Detection with Convolutional Neural Networks


Authors

Smruti Ranjan Swain Affiliation:
Gandhi Institute for Technology, Bhubaneshwar, India
, Biswadarsi Biswal Affiliation:
Assistant Professor, Dept. of CSE, Gandhi Institute for Technology, Bhubaneshwar, India
and Surabika Hota Affiliation:
Assistant Professor, Dept. of CSE, Gandhi Institute for Technology, Bhubaneshwar, India


Abstract

Psychological wellness conditions influence a noteworthy level of the world’s population every year. The stress investigation of emotional wellness phenomena in openly accessible social networking sites like Twitter,Sina Weibo and Facebook. A set of stress-related textual, visual, and social attributes from various aspects are first defined and then propose a novel hybrid model. The work has demonstrated the utility of online social information for contemplating despondency, be that as it may, there have been limited assessments of other mental wellbeing conditions. It is not easy to access the user posts on their Facebook page. In order to obtain the user data from Facebook, system have to get the access token from Facebook developer page. The API act as an intermediate system that will help the system to analysis the user information from the Facebook page. The system will also help to Recommending users with different links for psychological counselling centers, soft music or articles to help release their stress according to users’ stress level


Keywords

Stress detection, social media, microblog, access tokens, and face book.


Citation

Swain, S. R., Biswal, B., & Hota, S. (2019). A novel hybrid model for stress detection with convolutional neural networks. Turkish Journal of Computer and Mathematics Education, 10(3), 1286–1290.

Published by: Engineering Journals

Engineering Journals Logo