Keywords

Recommender Systems, Collaborative Filtering, Matrix Factorization, MovieLens, Cosine Similarity, RMSE, SVD, User-Item Interactions, Latent Factors, Personalization

Abstract

This study presents a comparative analysis of Collaborative Filtering and Matrix Factorization techniques applied to the MovieLens 100K dataset for building recommender systems. Collaborative Filtering, grounded in user similarity, and Matrix Factorization, which uncovers latent factors, are both critical in predicting user preferences. Our research involves the implementation and evaluation of these methods to determine their effectiveness in predicting movie ratings. Through meticulous experimentation, we optimized the models and assessed their performance using the Root Mean Square Error (RMSE) metric. Results indicate that both methods are robust, with Collaborative Filtering achieving a slightly better RMSE. This paper contributes to the field by providing insights into the accuracy of these prevalent recommender system algorithms and their practical application in enhancing user experience on movie platforms.

Semester

2026

Course Name

STA 6367 Data Science 2

Instructor Name

Dr. Rui Xie

College

College of Sciences

Included in

Data Science Commons

Share

COinS
 

Accessibility Statement

This item was created or digitized prior to April 24, 2027, or is a reproduction of legacy media created before that date. It is preserved in its original, unmodified state specifically for research, reference, or historical recordkeeping. In accordance with the ADA Title II Final Rule, the University Libraries provides accessible versions of archival materials upon request. To request an accommodation for this item, please submit an accessibility request form.