Link Prediction Model Based on the Topological Feature Learning for Complex Networks
- PDF / 2,092,328 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 48 Downloads / 186 Views
RESEARCH ARTICLE-COMPUTER ENGINEERING AND COMPUTER SCIENCE
Link Prediction Model Based on the Topological Feature Learning for Complex Networks Salam Jayachitra Devi1 · Buddha Singh1 Received: 3 October 2019 / Accepted: 30 April 2020 © King Fahd University of Petroleum & Minerals 2020
Abstract Link prediction tremendously gained concern in the field of machine learning by virtue of its real-world applicability on various fields including social network analysis, biomedicine, e-commerce, criminal activities, scientific community, etc. Several link prediction methods exist which are applicable to specific types of networks. Here, the primary aim of this paper is to perform feature extraction from the given real-time complex network using subgraph extraction technique and labeling of the vertices in the subgraph according to the distance from the vertex associated with each target link. The vertices in the subgraph are labeled based on the Geometric mean distance and Arithmetic mean distance. This proposed model helps to learn the topological pattern from the extracted subgraph. The feature extraction is carried out with different size of the subgraph with the number of vertices as K = 10 and K = 15. These features are then fit into different machine learning classification models and deep learning convolutional neural network model. For the evaluation purpose, area under the receiver operating characteristic curve (AUC) metric is used. The AUC results obtained from all the classifiers have been shown. Further, the simulation results show that bagging and random forest achieved good performance. Finally, the comparative study is performed to summarize the results and proved that link prediction using classification models and deep learning model perform well across different kinds of complex networks. This solved the link prediction problem with superior performance and with robustness. Keywords Link prediction · Classifiers · Deep learning · Topological feature · Complex networks
1 Introduction Recently, there has been an enormous increase in the availability of relational data. These data can be represented as social networks, scientific networks, biological networks [1], hybrid network [2–4], and transportation network [5–7]. Network analysis attracts increasing interest in the field of big data analysis and machine learning [8, 9]. Complex network analysis can also be performed with the concept of mathematical science, theoretical physics, and many more [10–12]. Link prediction is a fundamental task to perform an analysis of such relational data. The main purpose of link prediction is to predict future links or missing links * Salam Jayachitra Devi [email protected] Buddha Singh [email protected] 1
School of Computer and Systems Sciences, Jawaharlal Nehru University, New Delhi, India
among the entities in the network [1, 8, 13, 14]. It also has widespread applications including friend recommendation in social networks [15], product recommendation in a commercial network [16], protein to prote
Data Loading...
