Collaborative Filtering Recommendation Using Nonnegative Matrix Factorization in GPU-Accelerated Spark Platform

Abstract

Nonnegative matrix factorization (NMF) has been introduced as an efficient way to reduce the complexity of data compression and its capability of extracting highly interpretable parts from data sets, and it has also been applied to various fields, such as recommendations, image analysis, and text clustering. However, as the size of the matrix increases, the processing speed of nonnegative matrix factorization is very slow. To solve this problem, this paper proposes a parallel algorithm based on GPU for NMF in Spark platform, which makes full use of the advantages of in-memory computation mode and GPU acceleration. The new GPU-accelerated NMF on Spark platform is evaluated in a 4-node Spark heterogeneous cluster using Google Compute Engine by configuring each node a NVIDIA K80 CUDA device, and experimental results indicate that it is competitive in terms of computational time against the existing solutions on a variety of matrix orders. Furthermore, a GPU-accelerated NMF-based parallel collaborative filtering (CF) algorithm is also proposed, utilizing the advantages of data dimensionality reduction and feature extraction of NMF, as well as the multicore parallel computing mode of CUDA. Using real MovieLens data sets, experimental results have shown that the parallelization of NMF-based collaborative filtering on Spark platform effectively outperforms traditional user-based and item-based CF with a higher processing speed and higher recommendation accuracy.

Data Mining Technology Helps Digital Teaching and Learning of English Majors in Colleges and Universities

“…Non-negative matrix factorization (NMF) is a matrix factorization method that decomposes a given positive matrix into two positive matrices [20][21]. It is able to represent objects as non-negative linear combinations where partial information is extracted from a large number of objects.…”

Distributed Out-of-Memory NMF on CPU/GPU Architectures

“…Fig. 9: Results of Out of memory NMF benchmarks on Chicoma showing (a) NMF peak memory vs queue sizes for different k, and (b) NMF execution time vs queue sizes for different k.[32, 64, 128, 256, 512, 1024]. Smaller array H is cached on GPU memory, and large arrays A and W are stored on the host and batched to GPU as needed.…”