Welcome to Francis Academic Press

Academic Journal of Engineering and Technology Science, 2021, 4(2); doi: 10.25236/AJETS.2021.040215.

Implementation of Graph Embedding Algorithms in Node Classification for Wiki

Author(s)

Tong Wu, Ying Li

Corresponding Author:
Tong Wu
Affiliation(s)

New Media College, Communication University of Zhejiang, Hangzhou, China

Abstract

As wiki data integration has become one of the most popular and freely available data sources on the Internet, more and more domestic and foreign researchers are studying and analyzing wiki data in-depth, However, when classifying nodes in wiki data sets, we encountered poor relevance of node classification, which led to the problem of reduced data quality, accuracy, and reliability. In response to existing problems, we propose to use graph embedding related algorithms to classify wiki data sets. To improve the accuracy of the classification task, we have implemented Deepwalk (Deep Walk Random Algorithm), Node2vec, and SDNE (Structural Deep Network Embedding) The algorithm is tested on the wiki data set. The final experimental results show that the Graph Embedding algorithm has a good effect on the node classification of graph structure data, and the SDNE algorithm is the best inaccurate.

Keywords

Wiki dataset, Graph Embedding, Node Classification

Cite This Paper

Tong Wu, Ying Li. Implementation of Graph Embedding Algorithms in Node Classification for Wiki. Academic Journal of Engineering and Technology Science (2021) Vol. 4, Issue 2: 89-97. https://doi.org/10.25236/AJETS.2021.040215.

References

[1] Kou Leilei. Analysis of Data Sources in Wikidata. Library Theory and Practice. 2020:73-77

[2] Zhai Zhengli, Feng Shu, Li Penghui. Improved collaborative filtering recommendation model of graph convolutional network [J/OL]. Computer Engineering and Application: 1-10[2021-05-06]. http://kns.cnki.net/kcms/detail/11.2127.TP.20210419.1334.029.html.

[3] Liu Feng; Wang Baoliang; Zou Rongyu; Zhao Haochun. Research on the recommendation algorithm of network representation learning based on random walk. Computer Engineering: 1-9

[4] Grover Aditya;Leskovec Jure.node2vec: Scalable Feature Learning for Networks.[J].KDD: proceedings. International Conference on Knowledge Discovery & Data Mining, 2016, 2016:855-864.

[5] Ma Yang; Cheng Guangquan; Liang Xingxing; Li Yan; Yang Yuling; Liu Zhong. Improved SDNE algorithm in directed weighted network. Computer Science. 2020:239-243

[6] Xue Qiuhong, Jia Junzhi, Liu Huizhou. Implementation of semantic association between Chinese name specification data and Wikidata[J]. Information Theory and Practice, 2019, 42(10) :146-150.

[7] Zeng Ruming; Li Yunfei. Research on an improved method of K-means clustering algorithm. Journal of Shaoyang University (Natural Science Edition). 2021:14-20

[8] Xie Juanying, Qu Yanan. K-medoids clustering algorithm for density peak optimization initial center [J]. Computer Science and Exploration, 2016, 10(2) :230-247

[9] Han Linyi; Wu Sheng. Collaborative filtering algorithm based on K nearest neighbors optimized by scoring features. Information technology. 2018:83-87

[10] Bryan Perozzi,Rami Al-Rfou,Steven Skiena. DeepWalk[P]. Knowledge discovery and data mining, 2014.

[11] Grover Aditya,Leskovec Jure. node2vec: Scalable Feature Learning for Networks. [J]. KDD: proceedings. International Conference on Knowledge Discovery & Data Mining, 2016.

[12] Daixin Wang,Peng Cui,Wenwu Zhu. Structural Deep Network Embedding [P]. Knowledge Discovery and Data Mining, 2016.

[13] Wikidata main page[EB/OL].[2016-02-07].https://www.wikidata.org.