This is a recommendation system implemented in C++. The system utilizes a collaborative filtering algorithm to provide personalized recommendations based on user ratings. The code is designed to read input data from text files and output the recommendations in a CSV file.
- Collaborative filtering algorithm based on the Manhattan distance similarity metric.
- Predicts user ratings for items based on similar users' ratings.
- Customizable parameters for the number of users, items, and top user count.
- Input data is read from text files, and recommendations are written to a CSV file.
To run the recommendation system, you need:
- C++ compiler
- Development environment or IDE
- Clone the project repository:
$ git clone https://github.com/erogluegemen/Movie-Recommendation-System.git $ cd Movie-Recommendation-System - Open the source code file recommendation_system.cpp in a C++ development environment.
- Modify the necessary constants in the code according to your data.
- Build and run the code.
- Prepare your training and testing data files in the required format.
- Modify the code to provide the correct paths to your data files.
- Build and run the code.
- The system will generate a CSV file named submission.csv containing the recommendations.
The recommendation system uses a collaborative filtering algorithm based on the Manhattan distance similarity metric.
It calculates the similarity between the query user and other users in the dataset based on their rating profiles.
The top similar users are then selected, and their ratings are used to predict the rating for the query user and item.
If you encounter any problems, do not hesitate to contact.
@Egemen Eroglu
@Cagatay Tugcu
The recommendation system code was developed using C++ and the collaborative filtering algorithm. We would like to thank the contributors for their efforts in creating this system.
If you have any questions or need assistance, please don't hesitate to contact us.
Thank you for using the recommendation system!