Properties of rule interestingness measures and alternative approaches to normalization of measures

被引:39
作者
Greco, Salvatore [2 ]
Slowinski, Roman [1 ,3 ]
Szczech, Izabela [1 ]
机构
[1] Poznan Univ Tech, Inst Comp Sci, PL-60965 Poznan, Poland
[2] Univ Catania, Dept Econ & Business, I-95129 Catania, Italy
[3] Polish Acad Sci, Syst Res Inst, PL-01447 Warsaw, Poland
关键词
Rule interestingness measures; Properties of measures; Confirmation; Normalization; BAYESIAN CONFIRMATION MEASURES; ROUGH MEMBERSHIP; DECISION; SUPPORT;
D O I
10.1016/j.ins.2012.05.018
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We are considering properties of interestingness measures of rule; induced from data. These are: Bayesian confirmation property, two properties related to the case of entailment or refutation, called (Ex(1)) and logicality L, and a group of symmetry properties. We propose a modification of properties (Ex(1)) and L, called weak (Ex(1)), and weak 1, that deploy the concept of confirmation in its larger sense. We demonstrate that properties (Ex(1)) and L do not fully reflect such understanding of the confirmation concept, and thus, we propose to substitute (Ex(1)) by weak (Ex(1)) and L by weak L Moreover, we introduce four new approaches to normalization of confirmation measures in order to transform measures so that they would obtain desired properties. The analysis of the results of the normalizations of the confirmation measures takes into account all considered properties. We advocate for two normalized confirmation measures: measure Z considered in the literature, and newly proposed measure A. Finally, we provide some ideas for combining them in a single measure keeping all desirable properties. (C) 2012 Elsevier Inc. All rights reserved.
引用
收藏
页码:1 / 16
页数:16
相关论文
共 41 条
[1]  
Agrawal R., 1993, SIGMOD Record, V22, P207, DOI 10.1145/170036.170072
[2]  
[Anonymous], 1991, KNOWLEDGE DISCOVERY
[3]  
[Anonymous], 1962, LOGICAL FDN PROBABIL
[4]  
[Anonymous], 2007, Principles of Data Mining: Undergraduate Topics in Computer Science
[5]   Sequential covering rule induction algorithm for variable consistency rough set approaches [J].
Blaszczynski, Jerzy ;
Slowinski, Roman ;
Szelag, Marcin .
INFORMATION SCIENCES, 2011, 181 (05) :987-1002
[6]   Mining Pareto-optimal rules with respect to support and confirmation or support and anti-support [J].
Brzezinska, Izabela ;
Greco, Salvatore ;
Slowinski, Roman .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2007, 20 (05) :587-600
[7]   Measuring confirmation (Theory of rational belief) [J].
Christensen, D .
JOURNAL OF PHILOSOPHY, 1999, 96 (09) :437-461
[8]   On bayesian measures of evidential support: Theoretical and empirical issues [J].
Crupi, Vincenzo ;
Tentori, Katya ;
Gonzalez, Michel .
PHILOSOPHY OF SCIENCE, 2007, 74 (02) :229-252
[9]  
Eeills E, 2002, PHILOS STUD, V107, P129
[10]  
Fayyad UM, 1996, ADV KNOWLEDGE DISCOV, P1