id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-035033-osjy88rc	Aydin, Berkay	Spatiotemporal event sequence discovery without thresholds	2020-11-09		.txt	text/plain	8231	430	54	Here, we introduce a novel algorithm, RAND-ESMINER, which, by randomly repeating the mining process on a random subset of instances and follow relationships, finds an estimate participation index for event sequences. The RAND-ESMINER uses our pattern growth-based ESGROWTH algorithm [4] as the backbone, where the follow relationships are translated into a directed acyclic graph structure, and randomly permutes the edges of this graph to mine the event sequences. They defined a follow relation between the pointbased event instances of two different event types, presented significance measures for sequences, and introduced two pattern-growth based algorithms for the mining task. In this paper, we will focus on mining STESs using a randomization approach, which will take a set of spatiotemporal event instances as input and returns all the discovered STESs together with a list of estimated participation index values for each STES, obtained from randomized trials.	./cache/cord-035033-osjy88rc.txt	./txt/cord-035033-osjy88rc.txt