Hardware accelerator and method of collaborative filtering recommendation algorithm based on neighborhood model

Abstract

The invention discloses a hardware accelerator of a collaborative filtering recommendation algorithm based on a neighborhood model. The hardware accelerator comprises a training accelerator structure and a forecasting accelerator structure, wherein the training accelerator structure is used for accelerating a training stage of the collaborative filtering recommendation algorithm of the neighborhood model; the forecasting accelerator structure is used for accelerating a forecasting stage of the collaborative filtering recommendation algorithm of the neighborhood model; the training accelerator part can accelerate calculation of a Jaccard similarity coefficient, a Euclidean distance, two cosine similarities, a Pearson's correlation coefficient and an average difference degree which are involved in the training stage of the acceleration algorithm; and the forecasting accelerator part can calculate cumulative summing, weighting cumulative averaging and summing cumulative averaging which are involved in the forecasting stage of the acceleration algorithm. The hardware accelerator is good in acceleration effect and relatively low in power and energy consumption expense.

Description

technical field [0001] The invention relates to the field of computer hardware acceleration, in particular to a hardware accelerator and method for a neighborhood model-based collaborative filtering recommendation algorithm. Background technique [0002] The collaborative filtering recommendation algorithm based on the neighborhood model is a classic and mature algorithm in the field of recommendation algorithms, and is widely used in various recommendation systems, mainly including User-based CF algorithm, Item-based CF algorithm, and SlopeOne algorithm Wait. With the advent of the era of big data, the scale of data is growing rapidly, and the time it takes for this type of recommendation algorithm to process the ever-expanding data becomes longer and longer. In order to reduce the response time of the recommendation system and generate recommendation information for users in a timely manner, it is necessary to speed up the execution of this type of recommendation algorith...

AI Technical Summary

Problems solved by technology

[0004] Using the above three platforms to accelerate the relevant research work on this type of recommendation algorithm can indeed achieve good results, but there are also some problems: although the multi-core processor cluster and the cloud computing platform have a good acceleration effect on the whole, but The computing efficiency of a single computing node based on the GPP architecture is relatively low, and it has high energy consumption; although GPGPU has high computing efficiency, it also has the problems of high operating power and excessive energy consumption.

Images