Filters








2,000 Hits in 8.9 sec

Integrating classification capability and reliability in associative classification: A β-stronger model

Yuanchun Jiang, Yezheng Liu, Xiao Liu, Shanlin Yang
2010 Expert systems with applications  
We propose two new theorems to prune redundant frequent items and a concept of indiscernibility relationship between rules to prune redundant rules.  ...  Mining class association rules is an important task for associative classification and plays a key role in rule-based decision support systems.  ...  Discovery of class association rule undergoes two phases: mining complete rules from data set and extracting a minimal rule set by pruning the redundant ones.  ... 
doi:10.1016/j.eswa.2009.11.021 fatcat:jfgyml6evfdppbdilqyy4sjdja

Exploiting statistically significant dependent rules for associative classification

Jundong Li, Osmar R. Zaiane
2017 Intelligent Data Analysis  
In particular, we use Fisher's exact test as a significance measure to directly mine classification association rules by some effective pruning strategies.  ...  Without any threshold settings like minimum support and minimum confidence, SigDirect is able to find non-redundant classification association rules which express a statistically significant dependency  ...  Non-redundant CARs The CAR X → c k is non-redundant, if there does not exist any CARs in the form of Y → c k such that Y X and p F (Y → c k ) < p F (X → c k ). Definition 6.  ... 
doi:10.3233/ida-163141 fatcat:chyqpgc73vf7thtjmefvbjzh6y

Mining features for sequence classification

Neal Lesh, Mohammed J. Zaki, Mitsunori Ogihara
1999 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD '99  
We adapt data mining techniques to act as a preprocessor to select features for standard classification algorithms such as Naive Bayes and Winnow.  ...  We adapt data mining techniques to act as a preprocessor to select features for standard classi cation algorithms such as Naive B a yes and Winnow.  ...  We show below that we can reduce the number of features and the time needed to mine for features by pruning redundant rules.  ... 
doi:10.1145/312129.312275 dblp:conf/kdd/LeshZO99 fatcat:yzb7rbttbbajfekcy44rtvqytu

Assessing the statistical significance of association rules [article]

Wilhelmiina Hämäläinen
2014 arXiv   pre-print
In this paper, we inspect the most common measure functions - frequency, confidence, degree of dependence, χ^2, correlation coefficient, and J-measure - and redundancy reduction techniques.  ...  An association rule is statistically significant, if it has a small probability to occur by chance.  ...  Kolehmainen for checking the validity of statistical arguments and professor M. Nykänen for his valuable comments.  ... 
arXiv:1405.1360v1 fatcat:omk5abh4xbgedhs2efdcj5slvi

Efficient Discovery of the Most Interesting Associations

Geoffrey I. Webb, Jilles Vreeken
2013 ACM Transactions on Knowledge Discovery from Data  
This branch-and-bound algorithm deploys two powerful pruning mechanisms based on upper bounds on itemset value and statistical significance level.  ...  We present extensive evaluation of the strengths and limitations of the technique, including comparisons with alternative approaches to finding the most interesting associations.  ...  ACKNOWLEDGMENTS We are grateful to Tijl De Bie, Wilhelmiina H äm äl äinen, Michael Mampaey, Francois Petitjean, and Nikolaj Tatti for valuable comments and suggestions on drafts of this work.  ... 
doi:10.1145/2601433 fatcat:pvz4w2sqsjdtree5bvn5pxny6q

A general framework to encode heterogeneous information sources for contextual pattern mining

Weishan Dong, Wei Fan, Lei Shi, Changjin Zhou, Xifeng Yan
2012 Proceedings of the 21st ACM international conference on Information and knowledge management - CIKM '12  
We demonstrate by three applications of the contextual association rule, sequence and graph mining, that contextual patterns providing rich and insightful knowledge can be easily discovered by the proposed  ...  Traditional pattern mining methods usually work on single data sources. However, in practice, there are often multiple and heterogeneous information sources.  ...  Then frequent itemsets are enumerated and pruned as in typical generalized association rule mining.  ... 
doi:10.1145/2396761.2396774 dblp:conf/cikm/DongFSZY12 fatcat:4glau273ifgdja72qvhvimiqui

Finding semantic patterns in omics data using concept rule learning with an ontology-based refinement operator

František Malinka, Filip železný, Jiří Kléma
2020 BioData Mining  
The novel refinement operator uses two reduction procedures: Redundant Generalization and Redundant Non-potential, both of which help to dramatically prune the rule space and consequently, speed-up the  ...  These rules capture semantic differences between two classes: a target class as a collection of positive examples and a non-target class containing negative examples.  ...  R2, can be pruned from the set of rules R thus from the rule space R. Then the rules R1 and R2 are called Redundant Non-potentials.  ... 
doi:10.1186/s13040-020-00219-6 pmid:32905086 pmcid:PMC7466824 fatcat:fw2nl7nk2fffpdsc4vj4iypglu

What is Unequal among the Equals? Ranking Equivalent Rules from Gene Expression Data

Ruichu Cai, Anthony K. H. Tung, Zhenjie Zhang, Zhifeng Hao
2011 IEEE Transactions on Knowledge and Data Engineering  
In this paper, we look at two interestingness measures for ranking the interestingness of rules within equivalent rule group: Max-Subrule-Conf and Min-Subrule-Conf.  ...  In previous studies, association rules have been proven to be useful in classification problems over high dimensional gene expression data.  ...  All experiments below are done by repeating the 3-fold cross validation procedure [16] is repeated multiple times to obtain average readings.  ... 
doi:10.1109/tkde.2010.207 fatcat:wskrktuhdbb5pgln7iw75kdcuq

Integrated analysis of gene expression by Association Rules Discovery

Pedro Carmona-Saez, Monica Chagoyen, Andres Rodriguez, Oswaldo Trelles, Jose M Carazo, Alberto Pascual-Montano
2006 BMC Bioinformatics  
In this study we present a method for the integrative analysis of microarray data based on the Association Rules Discovery data mining technique.  ...  Microarray technology is generating huge amounts of data about the expression level of thousands of genes, or even whole genomes, across different experimental conditions.  ...  and by "UCM-Santander Central Hispano" through grant PR27/05-13964-BSCH.  ... 
doi:10.1186/1471-2105-7-54 pmid:16464256 pmcid:PMC1386712 fatcat:eukxnzloozdz3pr7ibtyk4sa2a

Sequence analysis and rule development of predicting protein stability change upon mutation using decision tree model

Liang-Tsung Huang, M. Michael Gromiha, Shinn-Ying Ho
2007 Journal of Molecular Modeling  
Therefore, iPTREE-2 based on a regression tree algorithm exhibits the ability of finding important factors and developing rules for the purpose of data mining.  ...  In the task of data mining, detailed analysis of sequences reveals the possibility of the compositional specificity of residues in different ranges of stability change and implies the existence of certain  ...  Moreover Oyama et al. proposed a data mining method to discover association rules related to protein-protein interactions [17] .  ... 
doi:10.1007/s00894-007-0197-4 pmid:17394029 fatcat:24gtz5ht55dc7lktuq4uqzrspi

Localization site prediction for membrane proteins by integrating rule and SVM classification

S. Zhou, K. Wang
2005 IEEE Transactions on Knowledge and Data Engineering  
The purpose of this paper is not improving the precision/recall of SVM, but is manifesting the rationale of a SVM classifier through partitioning the classification between if-then rules and the SVM classifier  ...  We study the localization prediction of membrane proteins for two families of medically important disease-causing bacteria, called Gram-Negative and Gram-Positive bacteria.  ...  Roughly speaking, almost 2/3 of features are redundant, 1/3 of non-redundant features have zero weight, and 1/3 of the remaining non-zero weight features are further pruned due to insignificance.  ... 
doi:10.1109/tkde.2005.201 fatcat:vnzisqdv6fdqbmertobavz63t4

Scalable feature mining for sequential data

N. Lesh, M.J. Zaki, M. Oglhara
2000 IEEE Intelligent Systems and their Applications  
As we show later, we can reduce the number of features and the time needed to mine for features by pruning redundant rules.  ...  The Rule-Prune procedure eliminates features based on our two pruning rules and based on length and width constraints.  ...  JUL/AUG -Corporate-Level Implementations of CMM Only a few have made it to Level 5. Who are they? Where have they succeeded? Where do they go from here? Is anything still missing?  ... 
doi:10.1109/5254.850827 fatcat:xadp7f7jsbdabevieqo7kkrcje

Mining Frequent Generalized Itemsets and Generalized Association Rules Without Redundancy

Daniel Kunkle, Donghui Zhang, Gene Cooperman
2008 Journal of Computer Science and Technology  
In particular, the two classification-based algorithms are MFGI class for mining max frequent g-itemsets and EGR class for mining essential g-rules.  ...  Our results fill an important gap among algorithms for frequent patterns and association rules by combining two concepts.  ...  Table 7 shows a sampling of generalized association rules found with a minimum support level of 0.1 and a minimum confidence level of 0.5.  ... 
doi:10.1007/s11390-008-9107-1 fatcat:x2o4f6yerjf7rhxcwzmcyvu2yq

ARUBAS: An Association Rule Based Similarity Framework for Associative Classifiers

Benoît Depaire, Koen Vanhoof, Geert Wets
2008 2008 IEEE International Conference on Data Mining Workshops  
The framework allows the researcher to use any association rule mining algorithm to produce the class association rules.  ...  In contrast with many existing associative classifiers, it uses class association rules to transform the feature space and uses instance-based reasoning to classify new instances.  ...  The algorithm dynamically sets the expected accuracy threshold during the mining process and also prunes redundant rules.  ... 
doi:10.1109/icdmw.2008.58 dblp:conf/icdm/DepaireVW08 fatcat:jezcv266ybdp7jopmdg4wlrjwu

Bridging Causal Relevance and Pattern Discriminability: Mining Emerging Patterns from High-Dimensional Data

Kui Yu, Wei Ding, Hao Wang, Xindong Wu
2013 IEEE Transactions on Knowledge and Data Engineering  
In order to address these two challenges, we bridge causal relevance and EP discriminability (the predictive ability of emerging patterns) to facilitate EP mining and propose a new framework of mining  ...  In this framework, we study the relationships between causal relevance in a causal Bayesian network and EP discriminability in EP mining, and then reduce the pattern space of EP mining to direct causes  ...  Based on CBA, Li et al. introduced CMAR (Classification based on Multiple-class AssociationRule) that generates classification association rules through a FP-tree and uses multiple rules to perform the  ... 
doi:10.1109/tkde.2012.218 fatcat:trohqshfjvghrnyvjakv3vtqnu
« Previous Showing results 1 — 15 out of 2,000 results