97,347 Hits in 7.9 sec

Rule Quality Measures for Rule Induction Systems: Description and Evaluation

Aijun An, Nick Cercone
2001 Computational intelligence  
A rule quality measure is important to a rule induction system for determining when to stop generalization or specialization.  ...  These formula-behavior rules are combined into formula-selection rules that can be used in a rule induction system to select a rule quality formula before rule induction.  ...  ACKNOWLEDGMENT The authors are members of the Institute for Robotics and Intelligent Systems (IRIS) and wish to acknowledge the support of the Networks of Centres of Excellence of the Government of Canada  ... 
doi:10.1111/0824-7935.00154 fatcat:eouxychnijbvtpliqgg5itltim

Rule Set Quality Measures For Inductive Learning Algorithms

Ralf Klinkenberg, Technische Universität Dortmund, Technische Universität Dortmund
This work provides several techniques for comparing and analyzing rule sets. These techniques measure the accuracy, generalization, time and space complexity, and domain coverage of rule sets.  ...  Symbolic inductive learning systems that induce concept descriptions from examples are valuable tools in the task of knowledge acquisition for expert systems.  ...  CONCLUSIONS This work proposed a set of measures to evaluate the quality of a rule set and to compare several rule sets generated by inductive learning algorithms.  ... 
doi:10.17877/de290r-7889 fatcat:eg6aaziuubc5rhbit4sown2ecm

The Structure of Description: Evaluating Descriptive Inferences and Conceptualizations

Marcus Kreuzer
2019 Perspectives on Politics  
The quality of description thus becomes a function of how well it addresses those challenges. I explicate distinct criteria for evaluating how well a describer handles those challenges.  ...  Explanation presumes description. Description explores the who, when, where, and how, and its answers furnish the raw material for theorizing and explaining.  ...  In historical description, four steps define this inference process and these steps also provide the basis for evaluating the quality of description.  ... 
doi:10.1017/s1537592718001068 fatcat:rmpp5grvlnaabffqjy2iixpbfm

Linguistic, Kinematic and Gaze Information in Task Descriptions: The LKG-Corpus

Tim Reinboth, Stephanie Gross, Laura Bishop, Brigitte Krenn
2020 International Conference on Language Resources and Evaluation  
To investigate this, we present the Linguistic, Kinematic and Gaze information in task descriptions Corpus (LKG-Corpus), comprising multimodal data on 13 humans, conducting take, put, and push actions,  ...  Recorded are audio, video, motion and eye-tracking data while participants perform an action and describe what they do.  ...  ), Ideen Lab 4.0 project CoBot Studio (872590), and the Austrian Science Fund (FWF), Project P29427-G24.  ... 
dblp:conf/lrec/ReinbothGBK20 fatcat:wpi47ssq4bb75akltczfqlkpsi

ELEM2: A learning system for more accurate classifications [chapter]

Aijun An, Nick Cercone
1998 Lecture Notes in Computer Science  
A new rule quality measure is proposed for the purpose of post-pruning. The measure is defined according to the relative distribution of a rule with respect to positive and negative examples.  ...  The method employs several new strategies in the induction and classification processes to improve the predictive performance of induced rules.  ...  Ning Shan for his suggestions on early versions of this work.  ... 
doi:10.1007/3-540-64575-6_68 fatcat:f73isj6wkfc2pmycrxbkgcsrnq

Learning two-tiered descriptions of flexible concepts: The POSEIDON system

F. Bergadano, S. Matwin, R. S. Michalski, J. Zhang
1992 Machine Learning  
In phase 2, this description is optimized accord-i~ng to a domain-dependent quality criterion.  ...  In phase 1, the AQ-15 rule learning program is applied to induce a complete and consistent concept description from supplied examples.  ...  Acknowledgments The authors express their gratitude to Hugo de Garis, Attilio Giordana, Ken Kaufman, Franz Oppacher, Lorenza Saitta, Gail Thornburg and Gheorghe Tecuci for many useful comments and criticisms  ... 
doi:10.1007/bf00994004 fatcat:ygya7a243ndshjmb7zo6nltcgu

Evolutionary Fuzzy Rule Induction Process for Subgroup Discovery: A Case Study in Marketing

MJ. del Jesus, P. Gonzalez, F. Herrera, M. Mesonero
2007 IEEE transactions on fuzzy systems  
This paper presents a genetic fuzzy system for the data mining task of subgroup discovery, the subgroup discovery iterative genetic algorithm (SDIGA), which obtains fuzzy rules for subgroup discovery in  ...  Experimental evaluation of the algorithm and a comparison with other subgroup discovery algorithms show the validity of the proposal.  ...  subgroup discovery systems, and a brief overview of GFSs for rule induction.  ... 
doi:10.1109/tfuzz.2006.890662 fatcat:ftho6r54vvhgtkzhn5vcevicv4

Rule Quality Measures Settings in a Sequential Covering Rule Induction Algorithm - an Empirical Approach

Marcin Michalak, Marek Sikora, Łukasz Wróbel
2015 Proceedings of the 2015 Federated Conference on Computer Science and Information Systems  
The paper presents the results of research related to the efficiency of the so called rule quality measures which are used to evaluate the quality of rules at each stage of the rule induction.  ...  The work is the continuation of research on the efficiency of quality measures employed in sequential covering rule induction algorithm.  ...  In the case of rule learning for descriptive purposes, the algorithms for induction of rules satisfying certain minimum quality criteria are most commonly used [12, 13, 14] .  ... 
doi:10.15439/2015f388 dblp:conf/fedcsis/MichalakSW15 fatcat:fhum64mvxrdbtp4k6aeno42rsu

An overview on subgroup discovery: foundations and applications

Franciso Herrera, Cristóbal José Carmona, Pedro González, María José del Jesus
2010 Knowledge and Information Systems  
An important characteristic of this task is the combination of predictive and descriptive induction. An overview related to the task of subgroup discovery is presented.  ...  Subgroup discovery is a data mining technique which extracts interesting rules with respect to a target variable.  ...  choice of the quality measures employed to extract and evaluate the rules.  ... 
doi:10.1007/s10115-010-0356-2 fatcat:7o6opddi4bfbpnutkghq7i335u

Rigel: An inductive learning system

Roberto Gemello, Franco Mana, Lorenza Saitta
1991 Machine Learning  
the use of rules to evaluate the worthiness of the inductive assertions.  ...  We assume the reader is acquainted with Induce and its terminology. However, for the sake of self-consistency, we give a brief description of AQ and Induce in the Appendix.  ...  Acknowledgments The authors are grateful to all the referees for their careful evaluation and many useful suggestions. Notes 1.  ... 
doi:10.1007/bf00153758 fatcat:jbvi4wcvgrcrvflgc7v7otm2ju

On Evaluating the Quality of Rule-Based Classification Systems

Nassim Dehouche
2017 ICIC Express Letters  
Two indicators are classically used to evaluate the quality of rule-based classification systems: predictive accuracy, i.e. the system's ability to successfully reproduce learning data and coverage, i.e  ...  . the proportion of possible cases for which the logical rules constituting the system apply.  ...  This article is the full version of an abstract entitled "On Evaluating the Quality of Machine Learning Classification Methods" presented at the Second International Conference on Mathematics and Statistics  ... 
doi:10.24507/icicel.11.10.1515 fatcat:s476i2xcjrastkjj2osjmfit3y

GPR: A Data Mining Tool Using Genetic Programming

Balasubramaniam Ramesh
2001 Communications of the Association for Information Systems  
We present GPR, an inductive data-mining system we developed. GPR uses the technique of genetic programming to discover rules.  ...  In the following section, we briefly define terminology and concepts related to knowledge discovery and the reasons for our focus on discovering production rules.  ...  This problem involves the search for inductive knowledge. The search for the set of high-quality descriptions can be viewed as an "optimization" problem.  ... 
doi:10.17705/1cais.00506 fatcat:xbklzxyinjgtfj7ifchudm4j74

Data-driven constructive induction

E. Bloedorn, R.S. Michalsi
1998 IEEE Intelligent Systems and their Applications  
The presented methodology concerns constructive induction, viewed generally as a process combining two intertwined searches: first for the "best" representation space, and second for the "best" hypothesis  ...  The aim of the search is to determine a generalized description of examples that optimizes a task-oriented multicriterion evaluation function.  ...  The complexity of a ruleset is evaluated by counting the number of rules in the ruleset and the total number of conditions (or selectors).The quality of a ruleset is evaluated lexicographically.  ... 
doi:10.1109/5254.671089 fatcat:g4oviuuty5hcppqd25jckk3wvu

NMEEF-SD: Non-dominated Multiobjective Evolutionary Algorithm for Extracting Fuzzy Rules in Subgroup Discovery

Cristóbal José Carmona, Pedro Gonzalez, María José del Jesus, Francisco Herrera
2010 IEEE transactions on fuzzy systems  
A study can be seen for the proposal and other previous methods for different databases.  ...  A new multi-objective evolutionary model for subgroup discovery with fuzzy rules is presented in this paper.  ...  The different measures used for evaluating a rule can be thought of as different objectives of the SD rule induction algorithm.  ... 
doi:10.1109/tfuzz.2010.2060200 fatcat:3gofpxkf4rdwpp6mxbrl2hoh4e

Automated Knowledge Acquisition: Overcoming the Expert System Bottleneck

David Perry Greene
1987 International Conference on Information Systems  
This paper describes the implementation of ADAM, a GA driven classifier, and compares the quality of the rules it generates to those of alternative induction techniques on a simulated decision problem.  ...  The artificial intelligence (AI) discipline of machine learning offers the best opportunity for alleviating the critical problem of acquiring the knowledge base necessary for expert systems.  ...  A brief description of ADAM can be found in the appendix; for a more mixed: decision = detailed description see Greene and Smith (1987) . 1 if ((Xl > t3) and (X2 + X3 +...  ... 
dblp:conf/icis/Greene87 fatcat:uwdy7ikyijeulbbptkyahfsmuq
« Previous Showing results 1 — 15 out of 97,347 results