Addressing Mark Imbalance in Integration-free Neural Marked Temporal Point Processes

TL;DR

This paper introduces a thresholding approach to address mark imbalance in neural marked temporal point processes, improving prediction accuracy for rare event marks without complex computations.

Contribution

It proposes a novel thresholding method combined with a neural MTPP model to handle mark imbalance and enhance prediction performance.

Findings

Outperforms baseline models in mark prediction accuracy

Effective time sampling without expensive numerical integration

Improved prediction for rare event marks

Abstract

Marked Temporal Point Process (MTPP) has been well studied to model the event distribution in marked event streams, which can be used to predict the mark and arrival time of the next event. However, existing studies overlook that the distribution of event marks is highly imbalanced in many real-world applications, with some marks being frequent but others rare. The imbalance poses a significant challenge to the performance of the next event prediction, especially for events of rare marks. To address this issue, we propose a thresholding method, which learns thresholds to tune the mark probability normalized by the mark's prior probability to optimize mark prediction, rather than predicting the mark directly based on the mark probability as in existing studies. In conjunction with this method, we predict the mark first and then the time. In particular, we develop a novel neural MTPP model to support effective time sampling and estimation of mark probability without computationally expensive numerical improper integration. Extensive experiments on real-world datasets demonstrate the superior performance of our solution against various baselines for the next event mark and time prediction. The code is available at https://github.com/undes1red/IFNMTPP.