Movie Recommendation System

(Data Mining Project)

Objective

• Build a Movie recommendation system based on “Association Rules”
• Use User Ratings data as the reference feature
• Create a website that gives Top 10 recommendations based on the movie selected

Approach

After thorough literature and technology survey, approach chosen for this project is as following:

• Use open-source Neflix user rating dataset from Kaggle (https://www.kaggle.com/datasets/netflix-inc/netflix-prize-data)
• Explore different FIM(Frequent Itemset Mining) Algorithms
• Compare all the models and choose best model

Project Steps:

• Data Collection
• Data Exploration
• Data Preprocessing
• Data Transformation
• Model Development
• Evaluation
• Live demo
• Applications and Future Scope

Contribution

Aneshaa Kasula, Viritha Vanama and Vishalakshi Ramanathan

This repository contains Jupyter notebooks focused on each stage of CRISP-DM such as data coolection, EDA visualizations,and so on using Python

References

• Yaman, Sezin & Fagerholm, Fabian & Munezero, Myriam & Männistö, Tomi & Mikkonen, Tommi. (2019). Patterns of User Involvement in Experiment-Driven Software Development. Information and Software Technology. 120. 106244. 10.1016/j.infsof.2019.106244.
• https://towardsdatascience.com/how-data-science-is-boosting-netflix-785a1cba7e45
• http://norma.ncirl.ie/4403/1/samruddhishaileshkanhere.pdf
• https://machinelearningknowledge.ai/best-explanation-of-apriori-algorithm-for-association-rule-mining/
• https://www.geeksforgeeks.org/depth-first-search-or-dfs-for-a-graph/
• https://www.youtube.com/watch?v=HV49RIUIx9Q
• https://www.youtube.com/watch?v=5ZOAJ6mn5Rg
• https://giphy.com/stickers/netflix-stream-streaming-binge-VD4svD2E2rMZXwuHkR