Improving Movie Recommendations Using Hybrid AI Systems: Leveraging Text-to-Number Conversion and Cosine Similarity

EasyChair Preprint 12700

Abstract

In this study, we propose an innovative approach to enhance movie recommendation systems through the integration of hybrid artificial intelligence (AI) techniques. Our method combines the power of text-to-number conversion and cosine similarity to improve the accuracy and relevance of movie recommendations. Text-to-number conversion allows us to transform textual data, such as user reviews or movie descriptions, into numerical representations, enabling efficient comparison and analysis. We leverage cosine similarity, a popular metric in information retrieval, to measure the similarity between movie features and user preferences. By integrating these techniques within a hybrid AI framework, we aim to provide personalized and contextually relevant movie recommendations to users. We evaluate the effectiveness of our approach through experiments on real-world movie datasets, demonstrating significant improvements in recommendation accuracy compared to traditional methods. Our findings suggest that hybrid AI systems, leveraging text-to-number conversion and cosine similarity, offer promising avenues for enhancing movie recommendation systems in practice.

Keyphrases: Artificial Intelligence, Information Retrieval, Movie Recommendations, Personalization, Text-to-number conversion, cosine similarity, hybrid systems

@booklet{EasyChair:12700,
  author    = {William Jack},
  title     = {Improving Movie Recommendations Using Hybrid AI Systems: Leveraging Text-to-Number Conversion and Cosine Similarity},
  howpublished = {EasyChair Preprint 12700},
  year      = {EasyChair, 2024}}