Research Article
Real-Time Movie Recommendation System using Locality-Sensitive Hashing
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357924, author={Baramkula Vishnu Pavan and D V S Mihir and Yogen Aralaguppi and Gayathri Ramasamy}, title={Real-Time Movie Recommendation System using Locality-Sensitive Hashing}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={locality-sensitive hashing (lsh) movie recommendation system collaborative filtering fuzzy matching minhash simhash cosine similarity resemblance similarity user preference dynamics collaborative filtering knn alternative}, doi={10.4108/eai.28-4-2025.2357924} }- Baramkula Vishnu Pavan
D V S Mihir
Yogen Aralaguppi
Gayathri Ramasamy
Year: 2025
Real-Time Movie Recommendation System using Locality-Sensitive Hashing
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357924
Abstract
The increasing volume of online movie content and the need to categorize and rate them requires an efficient recommendation system. To achieve this, Locality-Sensitive Hashing (LSH) and fuzzy matching (FM) are used in the recommendation system. The method of LSH is described, focusing on both user-based and item-based approaches. To achieve this, LSH schemes MinHash and SimHash are used, which use cosine similarity and resemblance similarity to compare data elements. Moreover, fuzzy matching techniques are used to accommodate the uncertainty in user ratings and preferences which improves the accuracy of recommendations. The movie recommendation system is trained on a real-world dataset. LSH and fuzzy matching algorithm have their advantages over other methods such as KNN and therefore are ideal for building real-time recommendation systems that can effectively address the challenges of big data and user preference dynamics.
