System and method for feature selection in decision trees

Details System and method for feature selection in decision trees System and method for feature selection in decision trees

2007-07-31

2007-07-31

Knight, Anthony (Department: 2121)

Data processing: artificial intelligence

Knowledge processing system

Reexamination Certificate

active

10185663

ABSTRACT:
Selection of certain attributes as output and input attributes is provided so a decision tree may be created more efficiently. For each possible output attribute an interestingness score is calculated. This interestingness score is based on entropy of the output attribute and a desirable entropy constant. The attributes with the highest interestingness score are used as output attributes in the creation of the decision tree. Score gains for the input attribute over the output attributes are calculated using a conventional scoring algorithm. The sum of the score gains over all output attributes for each input attribute is calculated. The attributes with the highest score gain sums are used as input attributes in the creation of the decision tree.

REFERENCES:
patent: 4852173 (1989-07-01), Bahl et al.
patent: 4872122 (1989-10-01), Altschuler et al.
patent: 6101275 (2000-08-01), Coppersmith et al.
patent: 6269353 (2001-07-01), Sethi et al.
patent: 6317735 (2001-11-01), Morimoto
patent: 6442561 (2002-08-01), Gehrke et al.
patent: 6643660 (2003-11-01), Miller et al.
patent: 6889219 (2005-05-01), Epstein et al.
patent: 6941287 (2005-09-01), Vaidyanathan et al.
patent: 6944329 (2005-09-01), Yoshii
patent: 6963870 (2005-11-01), Heckerman
patent: 7003490 (2006-02-01), Keyes
Almuallim, H. et al., “Development and Applications of Decision Trees”,Expert Systems, 2002, vol. 1, 53-77.
Fayyad, U.M. et al., “The Attribute selection problem in decision tree generation”,Proceedings of the 10thNational Conference on Artificial Intelligence, 1992, Cambridge MA, 104-110.
Giles, K. et al., “Comparasion of Two Families of Entropy-Based Classification Measures with and without Feature Selection”,34thAnnual Hawaii International Conference on System Sciences(HICSS-34) 2001, vol. 3, 3014.

Affiliated with

Also associated with

Rating

System and method for feature selection in decision trees does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with System and method for feature selection in decision trees, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for feature selection in decision trees will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3798974