Link Prediction Algorithm for Signed Social Networks Based on Local and Global Tightness
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Given that most of the link prediction algorithms for signed social networks can only complete sign prediction,a novel algorithm is proposed aiming to achieve both link prediction and sign prediction in signed networks. Based on the structural balance theory, the local link tightness and global link tightness are defined respectivelyby using the structural information of paths with the step size of 2 and 3 between the two nodes. Then the totalsimilarity of the node pair can be obtained by combining them. Its absolute value measures the possibility ofthe two nodes to establish a link, and its sign is the sign prediction result of the predicted link. The effectivenessand correctness of the proposed algorithm are verified on six typical datasets. Comparison and analysis are alsocarried out with the classical prediction algorithms in signed networks such as CN-Predict, ICN-Predict, andPSNBS (prediction in signed networks based on balance and similarity) using the evaluation indexes like areaunder the curve (AUC), Precision, improved AUC′, improved Accuracy′, and so on. Results show that theproposed algorithm achieves good performance in both link prediction and sign prediction, and its accuracy ishigher than other algorithms. Moreover, it can achieve a good balance between prediction accuracy andcomputational complexity.
Abstract
Given that most of the link prediction algorithms for signed social networks can only complete sign prediction,a novel algorithm is proposed aiming to achieve both link prediction and sign prediction in signed networks. Based on the structural balance theory, the local link tightness and global link tightness are defined respectivelyby using the structural information of paths with the step size of 2 and 3 between the two nodes. Then the totalsimilarity of the node pair can be obtained by combining them. Its absolute value measures the possibility ofthe two nodes to establish a link, and its sign is the sign prediction result of the predicted link. The effectivenessand correctness of the proposed algorithm are verified on six typical datasets. Comparison and analysis are alsocarried out with the classical prediction algorithms in signed networks such as CN-Predict, ICN-Predict, andPSNBS (prediction in signed networks based on balance and similarity) using the evaluation indexes like areaunder the curve (AUC), Precision, improved AUC′, improved Accuracy′, and so on. Results show that theproposed algorithm achieves good performance in both link prediction and sign prediction, and its accuracy ishigher than other algorithms. Moreover, it can achieve a good balance between prediction accuracy andcomputational complexity.
참고문헌 (10)
- M. M. Liu, "Link prediction in signed networks based on the similarity and structural balance theory", Journal of Information Hiding and multimedia Signal Processing, 8(4), pp.831~846 (2017)
- M. M. Liu, "Survey of link prediction algorithms in signed networks", Computer Science, 47(2), pp.21~30 (2020)
- D. Li, "Exploiting unlabeled ties for link prediction in incomplete signed networks", Proceedings of 2019 3rd IEEE International Conference on Robotic Computing (IRC), pp.538~543 (2019)
- X. Su, "Local labeling features and a prediction method for a signed network", CAAI Transactions on Intelligent Systems, 13(3), pp.437~444 (2018)
- P. Shen, "Unsupervised negative link prediction in signed social networks", Mathematical Problems in Engineering, 2019 (2019)[DOI]
- N. Girdhar, "Link prediction in signed social networks based on fuzzy computational model of trust and distrust", Soft Computing, 23(22), pp.12123~12138 (2019)
- X. Chen, "Link prediction in signed networks based on connection degree", Journal of Ambient Intelligence and Humanized Computing, 10(5), pp.1747~1757 (2019)
- X. Zhu, "Sign prediction on social networks based nodal features", Complexity, 2020 (2020)[DOI]
- G. Beigi, "Signed link prediction with sparse data : the role of personality information", Companion Proceedings of the 2019 World Wide Web Conference, pp.1270~1278 (2019)
- T. Derr, "Link and interaction polarity predictions in signed networks", Social Network Analysis and Mining, 10 (2020)[DOI]