id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-024461-xo75855r	Zhang, Yuanzhe	FCP Filter: A Dynamic Clustering-Prediction Framework for Customer Behavior	2020-04-17		.txt	text/plain	3883	227	55	In the meanwhile, Fragmentation and Coagulation Process (FCP), a stochastic partition model, has recently been proposed for identifying dynamic customer groups and modeling their purchase behavior. In our model, FCP clusters customers into groups by their temporal interests to filter random noise of individual transaction data. We conduct experiments on both synthetic and real-world datasets, demonstrating that our model is able to discover the latent group of individual customers and provides accurate predictions for dynamic purchase behavior. In order to track the customers' temporal shifting across groups, a novel Bayesian non-parametric customer segmentation model FC-CSM [7] based on a random partition process, Fragmentation and Coagulation Process (FCP) [1] , was proposed. We conducted experiments on synthetic and real-world datasets to illustrate that our model can (1) identify dynamic customer groups based on purchase behavior, (2) achieve more accurate prediction results by filtering individual random noise.	./cache/cord-024461-xo75855r.txt	./txt/cord-024461-xo75855r.txt