DM: FYI: preliminary results of missing values comparison

[Tjen-Sien Lim]

The following is the preliminary ranking of methods based on 16
datasets. Details for each dataset is available at
http://www.stat.wisc.edu/~limt/mv.html .


Methods                  Mean          Rank
                      Error Rate           
===========================================

Boosted See5 Rules      0.174            1
Boosted See5 Trees      0.176            2 
FSS Naive Bayes         0.195            3 
M5class                 0.197            4 
IND Bayes options       0.198            5 
IND CARTŪ  (0-SE)         0.199            6 
See5 Rules              0.199            7 
IND MML                 0.199            8 
IND MML options         0.200            9 
C4.5 Trees              0.201           10 
C4.5 Rules              0.202           11 
Naive Bayes             0.202           12 
See5 Trees              0.202           13 
IND CARTŪ  (1-SE)         0.203           14 
IND Bayes               0.205           15 
RPART                   0.208           16 
CARTŪ (tm)                0.208           17 
CN2                     0.210           18 
ITI                     0.212           19 
RIPPER                  0.214           20 
DMTI                    0.218           21 
KnowledgeSEEKER         0.219           22 
Boosted-RIPPER          0.225           23 
FIRM                    0.227           24

-- 
Tjen-Sien Lim                (608) 262-8181                        
Ph.D. candidate              limt@stat.wisc.edu                    
Dept. of Statistics          http://www.stat.wisc.edu/~limt
Univ. of Wisconsin-Madison
1210 West Dayton Street       
Madison, WI 53706