Mining Minimal Non-redundant Association Rules Using Frequent Closed Itemsets

Bastide, Yves; Pasquier, Nicolas; Taouil, Rafik; Stumme, Gerd; Lakhal, Lotfi

doi:10.1007/3-540-44957-4_65

Cited by 229 publications

(232 citation statements)

References 14 publications

Supporting

Mentioning

223

Contrasting

Unclassified

Order By: Relevance

“…Assume we are using a fixed window of size 6, and that, outside these occurrences, events a, b, c and d never occur. As in the previous example, 3 5 in s 1 and 1 5 in s 2 . Once again, the confidence values are not intuitive.…”

Section: Using Minimal Windowsmentioning

confidence: 58%

MARBLES: Mining Association Rules Buried in Long Event Sequences

Čule

Tatti

Goethals

2012

Proceedings of the 2012 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

Sequential pattern discovery is a well-studied field in data mining. Episodes are sequential patterns that describe events that often occur in the vicinity of each other. Episodes can impose restrictions on 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 and parallel episodes, while discovering general episodes is surprisingly understudied. This is particularly true when it comes to discovering association rules between them.In this paper we propose an algorithm that mines association rules between two general episodes. On top of the traditional definitions of frequency and confidence, we introduce two novel confidence measures for the rules. The major challenge in mining these association rules is pattern explosion. To limit the output, we aim to eliminate all redundant rules. We define the class of closed association rules, and show that this class contains all non-redundant output. To make the algorithm efficient, we use further pruning steps along the way. First of all, we generate only free and closed frequent episodes from which we create candidate rules, we speed up the evaluation of the rules, and finally prune the remaining non-closed rules from the output.

show abstract

Section: Using Minimal Windowsmentioning

confidence: 58%

MARBLES: Mining Association Rules Buried in Long Event Sequences

Čule

Tatti

Goethals

2012

Proceedings of the 2012 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

show abstract

“…From our point of view a resource could be a product, an application or any kind of service (e.g., multimedia, news and weather or connectivity infrastructure services). We have studied several measures from the literature to evaluate those which best fit in our recommender system, such as minimality [11,12], reliability [13], novelty [14], horting, predictability and closeness [10], and utility [5].…”

Section: Related Workmentioning

confidence: 99%

Resource Recommendation for Intelligent Environments Based on a Multi-aspect Metric

Almeida

Castillejo

López-de-Ipiña

et al. 2012

Ubiquitous Computing and Ambient Intelligence

View full text Add to dashboard Cite

Abstract. Intelligent environments offer information filled spaces. When trying to navigate among all the offered resources users can be overwhelmed. This problem is increased by the heterogeneous nature of resources in smart environments. Users must choose between a plethora of services, multimedia information, interaction modalities and devices. But at the same time the unique characteristics of smart spaces offers us more opportunities to filter these resources. To help users find the resource that they want and need we have designed a multi-aspect recommendation system that takes into account not only the features of the resource and the user, but also context data like the location and current activity. The developed system is flexible enough to be applied to different resource types and scenarios. In this paper we will describe the identified aspects and how they are merged into a single metric.

show abstract

“…In the last few years, the problem of efficiently mining closed patterns from a large r has attracted a lot of researchers in the data mining community (Bastide et al, 2000;Goethals & Zaki, 2003;Grahne & Zhu, 2003;Uno et al, 2004;Pan et al, 2003;Pasquier et al, 1999;Pei et al, 2000;Wang et al, 2003;Zaki & Hsiao, 2002). Let be a set of items, and be a transactional database defined on .…”

Section: Letmentioning

confidence: 99%

“…transactional databases (Bastide et al, 2000;Goethals & Zaki, 2003;Grahne & Zhu, 2003;Uno et al, 2004;Pan et al, 200 3;Pasquier et al, 1999;Pei et al, 2000;Wang et al, 2003;Zaki & Hsiao, 2002). The data structures are efficient and the mining speed is tremendously fast.…”

Section: Introductionmentioning

confidence: 99%

A Correspondence Between Maximal Complete Bipartite Subgraphs and Closed Patterns

Soh

et al. 2005

Lecture Notes in Computer Science

View full text Add to dashboard Cite

For an undirected graph ¢ without self-loop, we prove: (i) that the number of closed patterns in the adjacency matrix of ¢ is even; (ii) that the number of the closed patterns is precisely double the number of maximal complete bipartite subgraphs of ¢ ; (iii) that for every maximal complete bipartite subgraph, there always exists a unique and distinct pair of closed patterns that matches the two vertex sets of the subgraph. Therefore, we can efficiently enumerate all maximal complete bipartite subgraphs by using algorithms for mining closed patterns which have been extensively studied in the data mining field.

show abstract

Mining Minimal Non-redundant Association Rules Using Frequent Closed Itemsets

Cited by 229 publications

References 14 publications

MARBLES: Mining Association Rules Buried in Long Event Sequences

MARBLES: Mining Association Rules Buried in Long Event Sequences

Resource Recommendation for Intelligent Environments Based on a Multi-aspect Metric

A Correspondence Between Maximal Complete Bipartite Subgraphs and Closed Patterns

Contact Info

Product

Resources

About