Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Proactive Emotion Tracker: AI-Driven Continuous Mood and Emotion Monitoring

Mohammad Asif, Sudhakar Mishra, Ankush Sonker, Sanidhya Gupta, Somesh Kumar Maurya, Uma Shanker Tiwary

TL;DR

This paper presents a proactive emotion tracking framework that fuses social-media text analysis and browser history with physiological signals from wearables to detect depression and monitor mood. It leverages a modified pre-trained BERT model trained on Kaggle suicide-watch and Twitter datasets, achieving 93% test accuracy after targeted fine-tuning, surpassing traditional fine-tuning. The approach envisions long-term mood prognosis through EEG and wearable data, aiming for real-time monitoring and early intervention while addressing privacy with consent-based, local processing. The work offers a practical pathway toward privacy-conscious, multi-modal mood monitoring with potential applications in digital mental health care.

Abstract

This research project aims to tackle the growing mental health challenges in today's digital age. It employs a modified pre-trained BERT model to detect depressive text within social media and users' web browsing data, achieving an impressive 93% test accuracy. Simultaneously, the project aims to incorporate physiological signals from wearable devices, such as smartwatches and EEG sensors, to provide long-term tracking and prognosis of mood disorders and emotional states. This comprehensive approach holds promise for enhancing early detection of depression and advancing overall mental health outcomes.

Proactive Emotion Tracker: AI-Driven Continuous Mood and Emotion Monitoring

TL;DR

This paper presents a proactive emotion tracking framework that fuses social-media text analysis and browser history with physiological signals from wearables to detect depression and monitor mood. It leverages a modified pre-trained BERT model trained on Kaggle suicide-watch and Twitter datasets, achieving 93% test accuracy after targeted fine-tuning, surpassing traditional fine-tuning. The approach envisions long-term mood prognosis through EEG and wearable data, aiming for real-time monitoring and early intervention while addressing privacy with consent-based, local processing. The work offers a practical pathway toward privacy-conscious, multi-modal mood monitoring with potential applications in digital mental health care.

Abstract

This research project aims to tackle the growing mental health challenges in today's digital age. It employs a modified pre-trained BERT model to detect depressive text within social media and users' web browsing data, achieving an impressive 93% test accuracy. Simultaneously, the project aims to incorporate physiological signals from wearable devices, such as smartwatches and EEG sensors, to provide long-term tracking and prognosis of mood disorders and emotional states. This comprehensive approach holds promise for enhancing early detection of depression and advancing overall mental health outcomes.
Paper Structure (12 sections, 3 figures)

This paper contains 12 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Problem Statement
  3. Objectives
  4. Methodology
  5. Approach
  6. Dataset Creation and Model Training:
  7. Create a Script to Extract User Browser History:
  8. Datasets
  9. Model Architecture
  10. Fine Tuning
  11. Results
  12. Conclusion and Future Work

Figures (3)

  • Figure 1: An overall view of the work for social media and browsing history analysis for mood disorders detection.
  • Figure 2: Proposed model architecture of BERT Transformer.
  • Figure 3: Accuracy and Loss Graphs