The goal of this project is to integrate AI-based features into a yoga treatment website to improve its functioning and user experience.
Initially, LLM was OpenAI's GPT-4.0, whose responses were enhanced with Retrieval-Augmented Generation (RAG) using Vector Database (Marqo) to generate real and context-appropriate responses. However, we chose Perplexity's approach because of its improved capacity to provide more detailed and contextually rich information. Perplexity provides better contextual comprehension and more comprehensive reactions, making it an excellent choice for our specialist yoga treatment assistance.
To increase the accuracy and relevance of these responses, the bot searches a huge database of yoga-related content on PubMed and Google Scholar. Chainlit is used to deploy the bot and improve its usability. It also prompts users with follow-up questions, allowing them to refine their queries and receive more effective solutions. This interaction paradigm ensures that users obtain complete and accurate yoga treatment information that is personalized to their own needs.
Негізгі бет AI project - Yoga Therapy ChatBot
Пікірлер