Learning network event sequences using long short-term memory and second-order statistic loss
dc.contributor.author | Sha, Hao | |
dc.contributor.author | Al Hasan, Mohammad | |
dc.contributor.author | Mohler, George | |
dc.contributor.department | Computer and Information Science, School of Science | en_US |
dc.date.accessioned | 2022-05-09T14:43:07Z | |
dc.date.available | 2022-05-09T14:43:07Z | |
dc.date.issued | 2021-02 | |
dc.description.abstract | Modeling temporal event sequences on the vertices of a network is an important problem with widespread applications; examples include modeling influences in social networks, preventing crimes by modeling their space–time occurrences, and forecasting earthquakes. Existing solutions for this problem use a parametric approach, whose applicability is limited to event sequences following some well-known distributions, which is not true for many real life event datasets. To overcome this limitation, in this work, we propose a composite recurrent neural network model for learning events occurring in the vertices of a network over time. Our proposed model combines two long short-term memory units to capture base intensity and conditional intensity of an event sequence. We also introduce a second-order statistic loss that penalizes higher divergence between the generated and the target sequence's distribution of hop count distance of consecutive events. Given a sequence of vertices of a network in which an event has occurred, the proposed model predicts the vertex where the next event would most likely occur. Experimental results on synthetic and real-world datasets validate the superiority of our proposed model in comparison to various baseline methods. | en_US |
dc.eprint.version | Author's manuscript | en_US |
dc.identifier.citation | Sha, H., Al Hasan, M., & Mohler, G. (2021). Learning network event sequences using long short-term memory and second-order statistic loss. Statistical Analysis and Data Mining: The ASA Data Science Journal, 14(1), 61–73. https://doi.org/10.1002/sam.11489 | en_US |
dc.identifier.issn | 1932-1872 | en_US |
dc.identifier.uri | https://hdl.handle.net/1805/28872 | |
dc.language.iso | en | en_US |
dc.publisher | Wiley | en_US |
dc.relation.isversionof | 10.1002/sam.11489 | en_US |
dc.relation.journal | Statistical Analysis and Data Mining: The ASA Data Science Journal | en_US |
dc.rights | Publisher Policy | en_US |
dc.source | Author | en_US |
dc.subject | long short-term memory | en_US |
dc.subject | network-based events | en_US |
dc.subject | point processes | en_US |
dc.title | Learning network event sequences using long short-term memory and second-order statistic loss | en_US |
dc.type | Article | en_US |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- Sha2020Learning-AAM.pdf
- Size:
- 7.22 MB
- Format:
- Adobe Portable Document Format
- Description:
- Author's Manuscript
License bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- license.txt
- Size:
- 1.99 KB
- Format:
- Item-specific license agreed upon to submission
- Description: