Breadth-First Search Approach for Mining Serial Episodes with Simultaneous Events

Gandreti, SB and Ibrahim, A and Sastry, PS (2024) Breadth-First Search Approach for Mining Serial Episodes with Simultaneous Events. In: ACM International Conference Proceeding Series, 04-01-2024 to 07-02-2024, Bangalore, pp. 36-44.

PDF Acm_int_con_pro_ser_2024_pdf - Published Version Restricted to Registered users only Download (741kB) | Request a copy

Abstract

Frequent Episode Mining is a well-studied problem in the area of temporal data mining. There are many methods for mining serial, parallel and general partial order episodes. However, many of these existing methods are not very effective in capturing patterns where some events are constrained to occur simultaneously. There are a few methods for discovering such serial episodes; these methods use Depth-First Search based approaches and are not very efficient. In this paper, we propose a novel efficient algorithm for mining frequent serial episodes with simultaneous events. Our algorithm follows the Breadth-First Search approach, and, for this, we present a novel candidate generation method and formally prove its correctness. We also propose a small but significant modification to the traditional Finite State Automata based frequency counting which results in considerable speed-up of the frequency counting step. Through several simulation experiments involving both synthetic and real data, we demonstrate the efficiency of the proposed algorithm. © 2024 ACM.

Item Type:	Conference Paper
Publication:	ACM International Conference Proceeding Series
Publisher:	Association for Computing Machinery
Additional Information:	The copyright for this article belongs to Association for Computing Machinery.
Keywords:	Finite automata, Breadth-first-search; Complex event sequence; Complex events; Episode mining; Event sequence; Finite-state automata; Frequency counting; Frequent episode minings; Simultaneous event; Temporal data mining, Data mining
Department/Centre:	Division of Interdisciplinary Sciences > Management Studies
Date Deposited:	04 Mar 2024 07:02
Last Modified:	04 Mar 2024 07:02
URI:	https://eprints.iisc.ac.in/id/eprint/84175

Actions (login required)

View Item