• Please send your full manuscript to: jocm@vip.163.com

Personalized Recommendation Method of Power Information Operation and Maintenance Knowledge Based on Spark

Zhaoyang Qu 1, Pengfei Xu 1, Youxue Ren 2
1. School of Information Engineering of Northeast Dianli University, Jilin, China
2. Jilin Power Supply Company of Jilin Power Company, Jilin, China
Abstract—Power information operation and maintenance knowledge overload has become a pressing issue with the development of smart grid construction. The traditional personalized recommendation method cannot meet the demand of personalized recommendation of power information maintenance knowledge in big data environment. This paper proposes a method based on Spark which gives a personalized recommendation method of power information operation and maintenance knowledge. Firstly, an implicit rating mechanism is introduced, which can transform the learning behavior of users into implicit rating of power information operation and maintenance knowledge. Secondly, a personalized recommendation method combing knowledge features and user interests is designed. Finally, the personalized recommendation method, based on Spark, is applied to recommend power information operation and maintenance knowledge. The experimental results show that the method can effectively improve the accuracy and real-time of recommendation.

Index Terms—Spark, power information operation and maintenance knowledge, personalized recommendation, implicit rating, collaborative filtering

Cite: Zhaoyang Qu, Pengfei Xu, and Youxue Ren, “Personalized Recommendation Method of Power Information Operation and Maintenance Knowledge Based on Spark," Journal of Communications, vol. 11, no. 8, pp. 785-791, 2016. Doi: 10.12720/jcm.11.8.785-791
Copyright © 2013-2017 Journal of Communications, All Rights Reserved
E-mail: jcm@etpub.com