Health trauma and well-being assistant for Bengali seniors in household: a multimodal approach

Abstract

The increasing number of elderly individuals living alone has emerged as a pressing global concern. Our research aims to address this issue by developing advanced modules that can be integrated into a system that enhances the quality of life for older adults. The modules focus on medicine detection, fall detection, reminders for important tasks and events and providing companionship through friendly verbal interactions. Through the integration of cutting-edge deep learning techniques, diverse models and natural language processing (NLP), we have successfully designed an effective medication and well-being assistant. These modules use computer vision technology along with reinforced learning from human feedback and convolutional neural networks (CNNs) to reach our goal. The modules can be integrated into systems to empower elderly individuals to lead more active and fulfilling lives. Finally, this research contributes to the well-being and happiness of the elderly, highlighting the significance of comprehensive support systems in promoting their overall well-being.

Description

Cataloged from PDF version of thesis.
Includes bibliographical references (pages 58-60).
This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2024.

Keywords

Companion system, Deep learning, CNN, Reinforced learning, NLP, Medication habit

Citation

Endorsement

Review

Supplemented By

Referenced By