Developing Recommendation System Using Genetic Algorithm Based Alternative Least Squares
İclal ÖZCAN, Mete Çelik · 2018
In this study, a cloud based recommendation system was implemented. One of the aims of recommendation systems is to provide recommendation to users based on the records of other users. In recent, the suggestion systems are implemented in cloud computing due to the big data. The DataProc system is very useful for managing spark services in Google Cloud, one of the most popular cloud computing systems. One of the most used approaches in recommendation systems is Collaborative Filtering. In this study, a recommendation system was implemented by using Alternative Least Squares Method (ALS) with Collaborative Filtering in DataProc environment. The initial parameters of the ALS algorithm are optimized with the Genetic Algorithm (GA) to achieve more accurate estimates. The performance of the proposed system is tested on the movielens small dataset and the results are presented.