Global+Optimization+Algorithms+Theory+and+Application_Part20

Global+Optimization+Algorithms+Theory+and+Application_Part20...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 22.1 DATA-MINING-CUP 381 C1 31333333011130233 B 00000000000200022 A 33111333330332130 A C3 03331333011130231 B 30111233133033133 A 31133103011313123 B 02311333332332333 A 33011103011310123 B 10300321012202233 B 10023302313300100 N 13133032333113230 A 03213300330031031 N 03020000013303113 N 13331332003110200 N 23213331131003032 A 11000330203002300 N 03300220010030331 N 33113233330032133 A 31330333011330123 B 00203301133033010 N 01201323030333330 N 30223313301003001 B 30131230133013133 A 00113010002133100 B 30033000311103200 B 11121311103310003 A 11313132101000310 B 13312102313010013 A 31100222331222302 N 01333333011130230 B 31113333100312133 A 21313101111013100 B 00000000030200022 A 33111333330331133 A 31333333011130233 B 01201333030333310 N 00000000000200022 A 33111333330332133 A C2 03331333011130231 B 30111233133033133 A 31133103011313123 B 02311333332332333 A 33011103011310123 B 10023302313300100 N 13133032333113230 A 02232331022331121 B 11023300332213301 A 02311333332332333 A 03213300330031031 N 03020000013303113 N 13331332003110200 N 13331332003110200 N 03300220010030331 N 23213331131003032 A 03300220010000331 N 21130320011021302 A 33113233330032133 A 10023122212302322 A 11000330203002300 N 30210113033032112 N 11321310200313233 A 33113233330332133 A 31330333011330123 B 30223313301003002 B 00203301133033010 N 01201323030333330 N 30223313301003001 B 30131230133013133 A 00113010002133100 B 30033000311103200 B 11121311103310003 A 21133113001000202 B 11313132101000310 B 13312102313010013 A 01333333011130230 B 30223313301003002 B 31113333100312133 A 21313101111013100 B 11330302002121233 B 32021231303033130 A 00000000030200022 A 31133103011313123 B 13133032333113230 A 02311333332332333 A 21313101111013100 B 10030321130311103 A 33111330330332133 A C4 rule processing Figure 22.6: Some Pareto-optimal individuals among the evolved classifier systems. system. It consists of three rules which lead to a profit of 7222. C 2, with one additional rule, reaches 7403. The 31-rule classifier system C 3 provides a gain of 8748 to which the system with the highest profit evolved, C 4, adds only 45 to a total of 8793 with a trade-off of 18 additional rules (49 in total). As shown in Table 22.1 on page 377 , most feature values are or 1 and there are only very few 2 and 3-valued features. In order to find out how different treatment of those will influence the performance of the classifiers and of the evolutionary process, we slightly modified the condition semantics in Table 22.4 by changing the meaning of rule 2 from > 1 to ≤ 1 (compare with Table 22.2 on page 378 ). The progress of the evolution depicted in Figure 22.7 exhibits no significant difference to the first one illustrated in Figure 22.5 . With the modified rule semantics, the best classifier system evolved delivered a profit of 8666 by utilizing 37 rules. This result is also not very much different from the original version. Hence, the treatment of the features with the values 2 and 3 does not seem to have much influence on the overall result. In the first approach, 382...
View Full Document

Page1 / 20

Global+Optimization+Algorithms+Theory+and+Application_Part20...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online