Wu, X.: Rule induction with extension matrices (1998)
0.04
0.04291015 = product of:
0.1716406 = sum of:
0.1716406 = weight(_text_:handling in 3912) [ClassicSimilarity], result of:
0.1716406 = score(doc=3912,freq=2.0), product of:
0.4128091 = queryWeight, product of:
6.272122 = idf(docFreq=227, maxDocs=44421)
0.0658165 = queryNorm
0.41578686 = fieldWeight in 3912, product of:
1.4142135 = tf(freq=2.0), with freq of:
2.0 = termFreq=2.0
6.272122 = idf(docFreq=227, maxDocs=44421)
0.046875 = fieldNorm(doc=3912)
0.25 = coord(1/4)
- Abstract
- Presents a heuristic, attribute-based, noise-tolerant data mining program, HCV (Version 2.0), absed on the newly-developed extension matrix approach. Gives a simple example of attribute-based induction to show the difference between the rules in variable-valued logic produced by HCV, the decision tree generated by C4.5 and the decision tree's decompiled rules by C4.5 rules. Outlines the extension matrix approach for data mining. Describes the HCV algorithm in detail. Outlines techniques developed and implemented in the HCV program for noise handling and discretization of continuous domains respectively. Follows these with a performance comparison of HCV with famous ID3-like algorithms including C4.5 and C4.5 rules on a collection of standard databases including the famous MONK's problems
