New Modern Approach to Predict Users’ Sentiment Using CNN and BLSTM

  • R. Sathish Kumar Department of Computer Science and Engineering Manakula Vinayagar Institute of Technology, Kalitheerthalkuppam, Puducherry, India
Keywords: Sentimental Analysis, Recommendation System, Deep Learning, CNN, BLSTM, Social Networks

Abstract

In Today’s world social network play a vital role and provides relevant information on user opinion. This
paper presents emotional health monitoring system to detect stress and the user mood. Depending on results
the system will send happy, calm, relaxing or motivational messages to users with phycological disturbance. It also sends warning messages to authorized persons in case a depression disturbance is detected by monitoring system. This detection of sentence is performed through convolution neural network (CNN) and bi-directional long-term memory (BLSTM). This method reaches accuracy of 0.80 to detect depressed and stress users and also system consumes low memory, process and energy. We can do the future work of this project by also including the sarcastic sentences in the dataset. We can also predict the sarcastic data with the proposed algorithm 

Published
2022-10-06
How to Cite
[1]
R. S. Kumar, “New Modern Approach to Predict Users’ Sentiment Using CNN and BLSTM”, INJIISCOM, vol. 3, no. 2, pp. 183-194, Oct. 2022.