2026-05-06T12:55:45.587Z http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai

ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai:2:8882 2022-06-05T19:37:32Z 2

A Machine Learning approach for learning temporal point process 2022 http://ezaposleni.singidunum.ac.rs/rest/sciNaucniRezultati/oai/record/2/8882 A. Petrović A. Bisercic B. Delibasic D. Milenkovic Despite a vast application of temporal point processes in infectiousdisease diffusion forecasting, ecommerce, traffic prediction, preventivemaintenance, etc, there is no significant development in improving the simulationand prediction of temporal point processes in real-world environments. With thisproblem at hand, we propose a novel methodology for learning temporal pointprocesses based on one-dimensional numerical integration techniques. Thesetechniques are used for linearising the negative maximum likelihood (neML)function and enabling backpropagation of the neML derivatives. Our approach istested on two real-life datasets. Firstly, on high frequency point process data,(prediction of highway traffic) and secondly, on a very low frequency pointprocesses dataset, (prediction of ski injuries in ski resorts). Four different pointprocess baseline models were compared: second-order Polynomialinhomogeneous process, Hawkes process with exponential kernel, Gaussianprocess, and Poisson process. The results show the ability of the proposedmethodology to generalize on different datasets and illustrate how differentnumerical integration techniques and mathematical models influence the qualityof the obtained models. The presented methodology is not limited to thesedatasets and can be further used to optimize and predict other processes that arebased on temporal point processes (PDF) A machine learning approach for learning temporal point process. Available from: https://www.researchgate.net/publication/360596077_A_machine_learning_approach_for_learning_temporal_point_process [accessed Jun 05 2022]. article https://doi.org/10.2298/CSIS123456789X 1 10 1820-0214 Computer Science and Information Systems