Mining closed episodes with simultaneous events

Abstract

Sequential pattern discovery is a well-studied field in data mining. Episodes are sequential patterns describing events that often occur in the vicinity of each other. Episodes can impose restrictions to the order of the events, which makes them a versatile technique for describing complex patterns in the sequence. Most of the research on episodes deals with special cases such as serial, parallel, and injective episodes, while discovering general episodes is understudied.In this paper we extend the definition of an episode in order to be able to represent cases where events often occur simultaneously. We present an efficient and novel miner for discovering frequent and closed general episodes. Such a task presents unique challenges. Firstly, we cannot define closure based on frequency. We solve this by computing a more conservative closure that we use to reduce the search space and discover the closed episodes as a postprocessing step. Secondly, episodes are traditionally presented as directed acyclic graphs. We argue that this representation has drawbacks leading to redundancy in the output. We solve these drawbacks by defining a subset relationship in such a way that allows us to remove the redundant episodes. We demonstrate the efficiency of our algorithm and the need for using closed episodes empirically on synthetic and real-world datasets.

Efficiently Summarising Event Sequences with Rich Interleaving Patterns

“…Mining general patterns, patterns where the order of events are specified by a DAG is surprisingly hard. Even testing whether a sequence contains a pattern is NP-complete [18]. Consequently, research has focused on mining subclasses of episodes, such as, episodes with unique labels [1,12], strict episodes [19], and injective episodes [1].…”

Cohort Comparison of Event Sequences with Balanced Integration of Visual Analytics and Statistics

“…To show the results, Norén et al [43] used a graphical approach to visualize temporal associations. This work can be extended to address unique temporal constraints, such as dealing with concurrent events, which Cule et al address in the context of pattern mining [16] and association rule mining [50].…”

Statistical significance of episodes with general partial orders

“…The non-overlapped frequency for general partial order episodes satisfies the important property of anti-monotonicity unlike the minimal windows frequency (refer to [3] for an example). Also, currently there exist efficient algorithms for discovering episodes with general partial orders under this frequency measure [2,39,40]. In the rest of the paper, whenever we refer to frequency of an episode (unless otherwise mentioned), we mean non-overlapped frequency.…”

“…An apriori-based algorithm for mining frequent closed episode patterns is presented in [39]. A depth-first search based approach for mining closed episodes with general partial orders is presented in [40]. In this paper, we build on the method proposed in [2].…”