Structuration Sets with Implication Intensity

Gras, Régis; Briand, H.; Peter, Philippe

doi:10.1007/978-3-642-61159-9_13

Cited by 5 publications

(3 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For a rule R of the form R → R , it takes into account the implicative intensity between the rules of degree 1 composed of one attribute of R and one of R : ϕ g (R) = Max i≤k;j ≤h ϕ a i , a j hk . c R .c R 1/2 (8) with the same notations as definition 3.4. The major loop of the algorithm is bounded by the maximal height h max of the hierarchy.…”

Section: The Agglomerative Algorithmmentioning

confidence: 99%

“…The subjective aspect (userdriven) takes into account the prior knowledge and the goal of the users [19], and the objective aspect (data-driven) gives priority to statistical criteria [11]. In this paper, we focus on a statistical measure, the implicative intensity [7,8], which is an adaptation of the likelihood linkage coefficient [18] to the non-symmetrical relationships. Basically, this measure aims at expressing the surprise of a user faced with a new quasiimplication.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Discovering R-rules with a directed hierarchy

Gras

Kuntz

2005

Soft Comput

Self Cite

View full text Add to dashboard Cite

In this paper, we extend the classical notion of quasi-implication ("when a i is present then usually a j is also present") to R-rules (rules of rules), the premisses and the conclusions of which can be rules themselves. A new statistical measure, based on the implicative intensity defined by Gras for quasi-implications, is defined to assess the significance of R-rules on a data set. We show how to organize R-rules in a new combinatorial structure, the directed hierarchy, which is inspired by the classical hierarchical classification. An incremental algorithm is developed to find the most significant R-rule "amalgamation". An illustration is presented on a real data set stemming from a recent survey of the French Public Education Mathematical Teacher Society on the level in mathematics of pupils in the final year of secondary education and the perception of this subject.

show abstract

Section: The Agglomerative Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Discovering R-rules with a directed hierarchy

Gras

Kuntz

2005

Soft Comput

Self Cite

View full text Add to dashboard Cite

show abstract

“…3) The intensity of implication ϕ(X → Y ) measures the "surprise" of having few counter-examples for the X → Y rule as compared with a random law. This measure was introduced by Gras [12] to improve the evaluation of rule confidence. The basic idea of intensity of implication is to compare the number of counter-examples N (g(X), g(Y )) of the rule X → Y with the expected number N (U, U ) where U and U are two randomly selected subsets of O considered to be equals to respectively |g(X)| and |g(Y )|.…”

Section: Statistical Measures Of Rule Qualitymentioning

confidence: 99%

A User-Driven Process for Mining Association Rules

Kuntz

Guillet

Lehn

et al. 2000

Principles of Data Mining and Knowledge Discovery

View full text Add to dashboard Cite

This paper describes the components of a human-centered process for discovering association rules where the user is considered as a heuristic which drives the mining algorithms via a well-adapted interface. In this approach, inspired by experimental works on behaviors during a discovery stage, the rule extraction is dynamic : at each step, the user can focus on a subset of potentially interesting items and launch an algorithm for extracting the relevant associated rules according to statistical measures. The discovered rules are represented by a graph updated at each step, and the mining algorithm is an adaptation of the well-known A Priori algorithm where rules are computed locally. Experimental results on a real corpus built from marketing data illustrate the different steps of this process.

show abstract