Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Rule-based machine learning classification and knowledge discovery for complex problems
Urbanowicz R. ACM SIGEVOlution7 (2-3):3-11,2020.Type:Article
Date Reviewed: Dec 8 2022

The intricate spiraling problems in areas such as aeronautics, bioinformatics, and meteorology require innovative algorithms for intellectual discovery from unique data types and sources. How should “Space Age” intelligent-based systems be capitalizing on the reputable learning classifier systems (LCS)? To make the LCS suitable for resolving knotty real-life problems, Urbanowicz presents the interacting procedures, distinct qualities, and capabilities of ExSTraCS, an “extended supervised tracking and classifying system.”

The tenet of ExSTraCS is predicated on Michigan-style LCS--the rule-based machine learning algorithms for propagating learned patterns over a cooperative population of independently comprehensible classifiers. Like the Michigan-style LCS, ExSTraCS applies reiterative discovery and associates the inclusive search of genetic techniques with inward-looking optimization to corroborate or harmonize learning.

ExSTraCS is designed to lessen the imperatives about data and analytical models, reinforce regulated learning, provide adaptable algorithms for the analysis of diverse data types, and enhance robustness and ease of use. ExSTraCS is implemented in Python to (a) predictably recognize and learn from a mixture of discrete and continuous variables; (b) apply expert knowledge and rules to preprocess data; (c) retain enduring memory of attributes for algorithmic training and feedback; and (d) support rule shrinkage to expedite the detection of information.

The author visibly explains the “data pre-processing, algorithm learning/training, rule population post-processing” phases of ExSTraCS. He provides the details for installing ExSTraCS and performing a six-bit multiplexer analysis. The experimental results with the multiplexer show the easy interpretations of the solutions spread over populations.

The author persuasively presents the resources for providing clear-cut interpretations from rule-population metrics, significance testing, and the visualization of results in ExStraCS. Consequently, I strongly recommend the use of LCS and ExSTraCS for solving difficult problems.

Reviewer:  Amos Olagunju Review #: CR147520
Bookmark and Share
  Featured Reviewer  
 
Learning (I.2.6 )
 
 
Classifier Design And Evaluation (I.5.2 ... )
 
 
Knowledge Representation Formalisms And Methods (I.2.4 )
 
Would you recommend this review?
yes
no
Other reviews under "Learning": Date
Learning in parallel networks: simulating learning in a probabilistic system
Hinton G. (ed) BYTE 10(4): 265-273, 1985. Type: Article
Nov 1 1985
Macro-operators: a weak method for learning
Korf R. Artificial Intelligence 26(1): 35-77, 1985. Type: Article
Feb 1 1986
Inferring (mal) rules from pupils’ protocols
Sleeman D.  Progress in artificial intelligence (, Orsay, France,391985. Type: Proceedings
Dec 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2023 ThinkLoud®
Terms of Use
| Privacy Policy