Mining Episode Rules from Event Sequences Under Non-overlapping Frequency

Ouarem, Oualid; Nouioua, Farid; Fournier‐Viger, Philippe

doi:10.1007/978-3-030-79457-6_7

Cited by 7 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with classical SPM, repetitive SPM not only is more challenging, but also has many forms: general form (no condition) [33], disjoint [52], one-off [54], and nonoverlapping forms [55]. Note that the disjoint form was called the nonoverlapping form in some studies [34,46], which is far different from the nonoverlapping form in this study. To clarify the difference between disjoint and nonoverlapping, an illustrative example is shown as follows.…”

Section: Related Workmentioning

confidence: 84%

“…In the disjoint form [34,46], the first position of an occurrence is greater than the last position of its previous occurrence. Thus, there are two occurrences: <1,3,5> and <6,7,8>.…”

Section: Related Workmentioning

confidence: 99%

“…For example, pattern a[0, 3]d means that there are zero to three wildcards between a and d. Repetitions of patterns can be useful in terms of capturing valuable information from sequences, and gap constraint SPM therefore has many applications, such as septic shock prediction for ICU patients [29], keyphrase extraction [30], missing behaviors analysis [31], and pyramid scheme pattern mining [32]. However, gap constraint SPM is not only difficult to solve, but also has many forms, such as periodic gaps SPM [33], disjoint SPM [34], one-off SPM [35], and nonoverlapping SPM [23]. Previous research work has shown that the nonoverlapping SPM avoids the production of many redundant patterns [24]; for example, under the nonoverlapping condition, the pattern ad occurs twice in sequence adbdad.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MCoR-Miner: Maximal Co-Occurrence Nonoverlapping Sequential Rule Mining

Zhang

et al. 2023

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

The aim of sequential pattern mining (SPM) is to discover potentially useful information from a given sequence. Although various SPM methods have been investigated, most of these focus on mining all of the patterns. However, users sometimes want to mine patterns with the same specific prefix pattern, called co-occurrence pattern. Since sequential rule mining can make better use of the results of SPM, and obtain better recommendation performance, this paper addresses the issue of maximal co-occurrence nonoverlapping sequential rule (MCoR) mining and proposes the MCoR-Miner algorithm. To improve the efficiency of support calculation, MCoR-Miner employs depth-first search and backtracking strategies equipped with an indexing mechanism to avoid the use of sequential searching. To obviate useless support calculations for some sequences, MCoR-Miner adopts a filtering strategy to prune the sequences without the prefix pattern. To reduce the number of candidate patterns, MCoR-Miner applies the frequent item and binomial enumeration tree strategies. To avoid searching for the maximal rules through brute force, MCoR-Miner uses a screening strategy. To validate the performance of MCoR-Miner, eleven competitive algorithms were conducted on eight sequences. Our experimental results showed that MCoR-Miner outperformed other competitive algorithms, and yielded better recommendation performance than frequent co-occurrence pattern mining. All algorithms and datasets can be downloaded from https://github.com/wuc567/Pattern-Mining/tree/master/MCoR-Miner.

show abstract

Section: Related Workmentioning

confidence: 84%

“…In the disjoint form [34,46], the first position of an occurrence is greater than the last position of its previous occurrence. Thus, there are two occurrences: <1,3,5> and <6,7,8>.…”

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MCoR-Miner: Maximal Co-Occurrence Nonoverlapping Sequential Rule Mining

Zhang

et al. 2023

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

show abstract

“…• Occurrence-based frequency definitions: Those are definitions that track exactly the occurrences of episodes in the entire event sequence without initially subdividing the sequence into sub-sequences except in cases where the user applies a constraint on the size of occurrences (Achar et al, 2013) called the span or gap constraints. This category of definitions includes minimal occurrence-based frequency (Mannila & Toivonen, 1996), nonoverlapped occurrencebased frequency (Fournier-Viger et al, 2021;Ouarem et al, 2021), and distinct occurrence-based frequency (Achar et al, 2012a).…”

Section: Frequency Definitions In Femmentioning

confidence: 99%

A survey of episode mining

Ouarem,

Nouioua,

Fournier‐Viger

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Episode mining is a research area in data mining, where the aim is to discover interesting episodes, that is, subsequences of events, in an event sequence. The most popular episode‐mining task is frequent episode mining (FEM), which consists of identifying episodes that appear frequently in an event sequence, but this task has also been extended in various ways. It was shown that episode mining can reveal insightful patterns for numerous applications such as web stream analysis, network fault management, and cybersecurity, and that episodes can be useful for prediction. Episode mining is an active research area, and there have been numerous advances in the field over the last 25 years. However, due to the rapid evolution of the pattern mining field, there is no prior study that summarizes and gives a detailed overview of this field. The contribution of this article is to fill this gap by presenting an up‐to‐date survey that provides an introduction to episode mining and an overview of recent developments and research opportunities. This advanced review first gives an introduction to the field of episode mining and the first algorithms. Then, the main concepts used in these algorithms are explained. After that, several recent studies are reviewed that have addressed some limitations of these algorithms and proposed novel solutions to overcome them. Finally, the paper lists some possible extensions of the existing frameworks to mine more meaningful patterns and presents some possible orientations for future work that may contribute to the evolution of the episode mining field.This article is categorized under: Algorithmic Development > Spatial and Temporal Data Mining Algorithmic Development > Association Rules Technologies > Association Rules

show abstract

“…Recently Xiang et al proposed a big data algorithm for episode mining called LA-FRMH [46], but this algorithm did not handle the case of simultaneous events. Oualid et al proposed an algorithm called NONEPI [47] for episode rule mining using the concept of non-overlapping frequency. The primary goal of NONEPI is to find rules that are easier to interpret as occurrences must be non-overlapping, but this algorithm did not consider complex event sequences that are very common in real-world problems.…”

Section: Figure 1 An Illustration Of a Single-event Sequencementioning

confidence: 99%

A Scalable Analytical Framework for Complex Event Episode Mining With Various Domains Applications

2022

View full text Add to dashboard Cite

With the ubiquity of sensor networks and smart devices that continuously collect data, we face the challenge of analyzing the growing stream of data in real time. In recent years, there has been a huge need to gain useful knowledge by incrementally analyzing event sequence data. Although episode pattern mining techniques have existed for years, people have recently become more aware of their practical value in solving real-life domain problems such as manufacturing records, stock markets, and weather forecasts. The effective and efficient application of episode pattern mining techniques to analyze complex event data is becoming increasingly important for solving real-life problems in wide domains. However, few studies have focused on developing a scalable framework based on episode pattern mining of complex event sequences for applications in various domains. In this work, we propose a novel framework named SAAF (Scalable Analytical Application Framework) based on complex event episode mining techniques, including batch episode mining, delta episode mining, incremental episode mining, and pattern merging, to consider both efficiency and accuracy. Moreover, to enhance scalability, we adopt the lambda architecture with Apache Spark and Apache Spark Streaming as the system development framework. Finally, the experimental results on three real datasets of different domains and two benchmark datasets showed that the proposed SAAF framework exhibits excellent performance in terms of efficiency, accuracy, and scalability.

show abstract

Mining Episode Rules from Event Sequences Under Non-overlapping Frequency

Cited by 7 publications

References 13 publications

MCoR-Miner: Maximal Co-Occurrence Nonoverlapping Sequential Rule Mining

MCoR-Miner: Maximal Co-Occurrence Nonoverlapping Sequential Rule Mining

A survey of episode mining

A Scalable Analytical Framework for Complex Event Episode Mining With Various Domains Applications

Contact Info

Product

Resources

About