The MapReduce Parallel Study of KNN Algorithm

Abstract:

Although the parallelization KNN algorithm improves the classification efficiency of the algorithm, the calculation of the parallel algorithms increases with the increasing of the training sample data scale, affecting the classification efficiency of the algorithm. Aiming at this shortage, this paper will improve the original parallelization KNN algorithm in the MapReduce model, adding the text pretreatment process to improve the classification efficiency of the algorithm.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 2123-2127
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.2123
• Online since: July 2014
• Authors: Chun Ye Wang*, Xiao Feng Zhou
• Keywords: KNN Algorithm, MapReduce, Text Classification, Text Pretreatment

References

[1] Yinghui Yang. Research of Improvement to the Density-based Method for Reducing the Amount of Training Data and Application to kNN[D]. Chongqing University, 2010. In Chinese.

[2] Si Wen. The Research of K-Nearest Neighbor Classification Algorithm Based on Hadoop Platform[D]. South China University of Technology, 2011. In Chinese.

[3] Hongxin Qiao. Research and Application of KNN Classification Algorithm Based on MapReduce[D]. Beijing Jiaotong University, 2012. In Chinese.

[4] Yihong Hong. Network and Information, 2010, 09: 36-37. In Chinese.

[5] Biao Liu. The Research of a MapReduce-Based kNN Algorithm in Spatial Database[D]. Dalian Maritime University, 2012. In Chinese.

[6] Yonggang Yan, Tinghuai Ma, Jian Wang. Transactions of Nanjing University of Aeronautics & Astronautics, 2013, 04: 550-555. In Chinese.

[7] Guoying Yang. Research on Text Classification Algorithm Based on MapReduce Model[D]. Liaoning University, 2013. In Chinese.