Affiliation:
1. Department of Computer Science and Technology, Tsinghua University
2. DAMO Academy, Alibaba Group
3. Ant Financial
Abstract
Analyzing large-scale evolving graphs are crucial for understanding the dynamic and evolutionary nature of social networks. Most existing works focus on discovering repeated and consistent temporal patterns, however, such patterns cannot fully explain the complexity observed in dynamic networks. For example, in recommendation scenarios, users sometimes purchase products on a whim during a window shopping.Thus, in this paper, we design and implement a novel framework called BurstGraph which can capture both recurrent and consistent patterns, and especially unexpected bursty network changes. The performance of the proposed algorithm is demonstrated on both a simulated dataset and a world-leading E-Commerce company dataset, showing that they are able to discriminate recurrent events from extremely bursty events in terms of action propensity.
Publisher
International Joint Conferences on Artificial Intelligence Organization
Cited by
7 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. ForeSeer: Product Aspect Forecasting Using Temporal Graph Embedding;Proceedings of the 32nd ACM International Conference on Information and Knowledge Management;2023-10-21
2. A Hybrid Continuous-Time Dynamic Graph Representation Learning Model by Exploring Both Temporal and Repetitive Information;ACM Transactions on Knowledge Discovery from Data;2023-06-15
3. TEA: A General-Purpose Temporal Graph Random Walk Engine;Proceedings of the Eighteenth European Conference on Computer Systems;2023-05-08
4. Temporal link prediction based on node dynamics;Chaos, Solitons & Fractals;2023-05
5. TeGraph: A Novel General-Purpose Temporal Graph Computing Engine;2022 IEEE 38th International Conference on Data Engineering (ICDE);2022-05