The aim of change-point detection is to discover the changes in behavior that lie behind time sequence data. In this article, we study the case where the data comes from an inhomogeneous Poisson process or a marked Poisson process. We present a methodology for detecting multiple offline change-points based on a minimum contrast estimator. In particular, we explain how to handle the continuous nature of the process with the available discrete observations. In addition, we select the appropriate number of regimes via a cross-validation procedure which is really handy here due to the nature of the Poisson process. Through experiments on simulated and real data sets, we demonstrate the interest of the proposed method. The proposed method has been implemented in the R package \texttt{CptPointProcess} R.

翻译：变点检测旨在发现时间序列数据背后的行为变化。本文研究数据来自非齐次泊松过程或标记泊松过程的情形。我们提出了一种基于最小对比估计量的方法，用于检测多个离线变点。具体而言，我们阐述了如何利用已有的离散观测值来处理过程的连续性。此外，我们通过交叉验证过程选择适当的区制数量，由于泊松过程的特性，该方法在此处非常实用。通过在模拟数据集和真实数据集上的实验，我们展示了所提方法的有效性。该方法已在R语言包\texttt{CptPointProcess}中实现。