Komorebi: Enhancing Mental Wellness with Sentiment Analysis and Cognitive Behavior Therapy

Abstract

In recent years, the rapid advancements in human and computer interaction have been reshaped by artificial intelligence (AI) and natural language processing (NLP). Chatbots have been among the most prominent use cases of NLP and have provided transformative technological applications which have been integral to our digital landscape. One of the most promising applications of chatbot utilization is in the medical industry, specifically mental health. NLP is at the heart of this AI capability and has revolutionized how chatbots interact with users, allowing the computer to generate, understand, and analyze human language interactions. This research presents the development and evaluation of a GPT-powered mental health chatbot for a mental wellness website with a foundation built on sentiment analysis techniques. The chatbot aims to provide personalized support and leverage evidence-based strategies for coping with and improving mental health issues prescribed using cognitive behavior therapy (CBT) interventions. Through the implementation of a combination of machine learning models and rule-based algorithms, the website features an activities module which categorizes user sentiments into positive, neutral, and negative emotions and prescribes a personalized CBT intervention to users. The system fosters an empathetic and supportive conversational environment between users and the virtual assistant to build rapport and trust. The effectiveness of this chatbot is measured through a mixed-methods approach that includes functional testing, user feedback, and ethical evaluation.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.