This project examines the efficacy of Large Language Models (LLMs) integrated with parallel computing to optimize recommendation systems. With a focus on Retrieval Augmented Generation (RAG) models, the research assesses the impact of parallel processing on the speed and relevance of book recommendations. Methodologically, the study contrasts the performance of traditional sequential processing against a parallelized approach across key operations such as data preprocessing, embedding generation, similarity computation, and recommendation prediction. Results indicate that parallel computing significantly decreases operational times by up to 92.89%, enhancing the system's efficiency. The practical application of the RAG model in a chatbot interface confirms the model's capability to deliver personalized and contextually appropriate book suggestions. These findings highlight the potential of integrating parallel processing with LLMs to advance the responsiveness and accuracy of content recommendation systems.
- Parallel Computing: Enhances the efficiency of data processing operations, significantly reducing the time required for tasks such as data preprocessing, embedding generation, and similarity computation.
- RAG Model: Utilizes Retrieval Augmented Generation to deliver personalized and contextually appropriate book recommendations.
- Scalability: Designed to handle large datasets efficiently through parallel processing techniques.
- Chatbot Integration: Provides a practical application of the recommendation system in a chatbot interface.