The modern day internet faces a very famous problem called information overload. Where the amount of information is huge and the need for personalized results to match ones preferences for ease of access to other information like it. This is especially a problem in the e-commerce and streaming industries where the amount of items available is massive and users need a way to surf through results quickly and efficiently to find the exact items they are looking for and possibly look at similar recommendations. Modern day recommendation engines use user-item data to find items an active user may like based on other users with similar preferences and provide recommendations. This paper looks at a model based approach, specifically collaborative filtering, to providing accurate recommendations. The model will be made based on normal predictor, singular vector decomposition, k-nearest neighbour, and slope one and the performance and accuracy of the models will be compared against each other to see the comparison between them.
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research (Dubai)
Ahmed, Shaheer Airaj, "Building a User-Based Recommendation System Using a Model-Based Collaborative Filtering Approach" (2021). Thesis. Rochester Institute of Technology. Accessed from