Web proceedings papers


Efthalia Karydi , Konstantinos Margaritis and Eero Vainikko


One of the problems that are encountered in recommender systems applications is the high sparsity of the available data. In this paper we investigate the e ect of the sparsity of datasets to the performance of a parallel implementation of the Collaborative Filtering Slope One algorithm. To represent the sparse data the Compressed Sparse Row (CSR) format is used and the implementation's performance is evaluated on a Graphics Processing Unit using the MovieLens and arti cially created datasets.


Collaborative Filtering, Slope One, CSR Format, Massively Parallel Computing, GPU, CUDA