PERSONALIZED E-COMMERCE RECOMMENDATION SYSTEM BASED ON COLLABORATIVE FILTERING UNDER HADOOP
Journal Title: Topics in Intelligent Computing and Industry Design (ICID) - Year 2017, Vol 1, Issue 2
Abstract
Aiming at the disadvantages of the traditional collaborative filtering recommendation algorithm based on user rating matrix in terms of items recommendation accuracy and scalability, a personalized e-commerce recommendation system based on Hadoop is designed and implemented. The data sparseness of user-item rating matrix leads to great contingency of recommendation results, and the more sparseness scoring matrix, the worse the recommended effect. The system in this paper uses the e-commerce platform implicit user rating and explicit rating data to the user preference characteristics as the impact factor to construct and replace the rate matrix. Finally, with the modified user-item interaction information, the Top-N recommendation method is adopted to recommend the products for the target user may be interest in. Comparative experimental results show that the recommendation system can improve the recommendation accuracy of products. The experimental results on the Hadoop platform also show that using the MapReduce parallel computing framework can improve computational eff iciency and algorithm scalability.
Authors and Affiliations
Haodong Tang, Xiaohui Cheng
DESIGN AND FABRICATION OF PHASE – LOCKED LOOP FREQUENCY SYNTHESIZER CIRCUIT
This paper introduces the design and fabrication of phase-locked loop frequency synthesizer circuit based on ADF4351 chip. System master chip selection STM32F103C8T6, power chip selection AMS1117-3.3V, display part of th...
THE INNOVATIVE DESIGNING OF THE CONNECTING-ROD SWINGROD GEARDRIVING INTERNAL-COMBUSTION ENGINES
Aiming at the current problems like short drive arms and low mechanism efficiency existed in the crank connecting rod piston internal-combustion engines, the paper proposes a design of a kind of connecting-rod swing-rod...
CHARACTERIZATION OF THE UV – VISIBLE ABSORPTION SPECTRA OF COMMONLY USED PHOTOINITIATORS
UV-LED curing has the advantages of energy saving and no ozone, mercury pollution has become one of the recently developed green printing technology.The photoinitiator used to meet the requirements of food packaging mate...
SALIENCY DETECTION VIA DENSE CONVOLUTION NETWORK
Saliency detection is a fundamental problem in the field of image processing and computer vision. The convolutional model has been also used in saliency detection for its outstanding performance on image classification a...
ON THE INFLUENCE OF CHINA-ENGLISH BY THE EXAMPLE OF KUNG FU PANDA’S SUBTITLE TRANSLATION
In the translation of the mass culture media of the film, the use of “China-English” can easier achieve the purpose of cultural propaganda and communication. This article mainly focuses on the analysis of subtitles trans...