An effective framework for monitoring Depression (A Mental Disorder) using Sentiment Analysis and Affective Computing

Abstract

The technique efficiently tracks depression, a mental illness, using sentiment analysis and affective computing. It offers reliable and prompt identification, beneficial assistance, and interventions for at-risk people by analyzing reactions and communication in digital situations. The framework's drawback is its reliance on textual data processing, which could ignore invisible indicators and variation regarding the way depression symptoms are expressed, reducing accuracy in some situations. In this paper, we offer Attribute-attention based LSTM (AALSTM) for strengthening an effective framework for monitoring depression to overcome this crucial issue. Initially, we gather the tweets dataset and preprocess the collected data to remove duplicates and guarantee homogeneity. The subsequent stage involves extracting pertinent features from the pre-processed data. We simulate trials with Python 3.11 software to assess the efficiency of the suggested algorithm. In terms of accuracy (99.53%), precision (99.58%), recall (99.51%), and F-Measure (99.56%), our results show that the AALSTM technique outperforms other methods in effectiveness. Our suggested AALSTM technique provides exciting outcomes for securing an effective framework for monitoring depression using Sentiment Analysis and Affective Computing.