ABSTRACT
We introduce a new ground-truth recovery boosting approach on antagonistic networks using the status-influence space obtained by the frustration cloud --- a generalization of the frustration index which models nearest consensus-based states of a signed graph. A spectral clustering and k-means approach are both examined and are compared to existing clustering methodologies on two sentiment-based datasets. We demonstrate that our approach successfully recovers all community labels on a highly modular dataset and outperforms the leading clustering technique by a factor of 3.08 on a more complex network. Additionally, we demonstrate that status and influence, in combination with network data, can be used to detect and characterize anomalous outcomes in promotion networks.
- K-Y. Chiang, J.J. Whang, and I. S. Dhillon. Scalable clustering of signed networks using balance normalized cut. In Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12, page 615, Maui, Hawaii, USA, 2012. ACM Press.Google ScholarDigital Library
- M. Cucuringu, P. Davies, A. Glielmo, and H. Tyagi. SPONGE: A generalized eigenproblem for clustering signed networks. arXiv:1904.08575 [cs, math, stat], 2019.Google Scholar
- F. Harary. On the notion of balance of a signed graph. Michigan Math. J., 2:143--146, 1953.Google ScholarCross Ref
- Jérôme Kunegis, Andreas Lommatzsch, and Christian Bauckhage. The slashdot zoo: Mining a social network with negative edges. In Proceedings of the 18th International Conference on World Wide Web, WWW '09. ACM, 2009.Google ScholarDigital Library
- Jure Leskovec and Andrej Krevl. SNAP Datasets: Stanford large network dataset collection. http://snap.stanford.edu/data, June 2014.Google Scholar
- Pedro Mercado, Francesco Tudisco, and Matthias Hein. Clustering signed networks with the geometric mean of laplacians. Advances in Neural Information Processing Systems, 12 2016.Google Scholar
- Pedro Mercado, Francesco Tudisco, and Matthias Hein. Spectral Clustering of Signed Graphs via Matrix Power Means. In Proceedings of the 36th International Conference on Machine Learning, pages 4526--4536, 2019.Google Scholar
- Kenneth Read. Cultures of the central highlands, new guinea. Southwestern Journal of Anthropology, 10(1):1--43, 1954.Google ScholarCross Ref
- Lucas Rusnak and Jelena Tešić. Characterizing attitudinal network graphs through frustration cloud. Data Mining and Knowledge Discovery, 6, November 2021.Google ScholarDigital Library
- S. Sampson. A novitiate in a period of change: An experimental and case study of relationships. Ph.D. Thesis, 1968.Google Scholar
Index Terms
- Cluster boosting and data discovery in social networks
Recommendations
An Artificial Bee Colony ABC Algorithm for Efficient Partitioning of Social Networks
Social networks clustering is an NP-hard problem because it is difficult to find the communities in a reasonable time; therefore, the solutions are based on heuristics. Social networks clustering aims to collect people with common interest in one group. ...
Detecting Overlapping Communities in Social Networks Using A Modified Segmentation by Weighted Aggregation Approach
DATA'21: International Conference on Data Science, E-learning and Information Systems 2021Detecting communities of common behaviors, interests, and interactions in social networks is essential to model a network's structure. Overlapping community detection is an NP-Hard problem. Several solutions have been proposed; however, most of these ...
Organizational overlap on social networks and its applications
WWW '13: Proceedings of the 22nd international conference on World Wide WebOnline social networks have become important for networking, communication, sharing, and discovery. A considerable challenge these networks face is the fact that an online social network is partially observed because two individuals might know each ...
Comments