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{\textendash}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{\textquoteright}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.
}, year = {2020}, journal = {Statistical Analysis and Data Mining: The ASA Data Science Journal}, volume = {14}, month = {11}, publisher = {Wiley Blackwell (John Wiley \& Sons)}, issn = {1932-1864}, url = {https://par.nsf.gov/biblio/10382535}, doi = {10.1002/sam.11489}, }