Como analisar o resultado do algoritmo REPTree do WEKA?

=== Run information ===

Scheme: weka.classifiers.trees.REPTree -M 2 -V 0.001 -N 3 -S 1 -L -1 -I 0.0
Relation: pima_diabetes
Instances: 768
Attributes: 9
preg
plas
pres
skin
insu
mass
pedi
age
class
Test mode: evaluate on training data

=== Classifier model (full training set) ===

REPTree

plas < 139.5
|   mass < 26.3 : tested_negative (101/3) [44/2]
|   mass >= 26.3
|   |   plas < 94.5 : tested_negative (65/3) [42/8]
|   |   plas >= 94.5
|   |   |   pedi < 0.53
|   |   |   |   age < 27.5 : tested_negative (62/9) [29/6]
|   |   |   |   age >= 27.5
|   |   |   |   |   age < 53.5
|   |   |   |   |   |   pedi < 0.14 : tested_negative (6/0) [0/0]
|   |   |   |   |   |   pedi >= 0.14
|   |   |   |   |   |   |   mass < 43.1
|   |   |   |   |   |   |   |   pres < 85
|   |   |   |   |   |   |   |   |   pedi < 0.49
|   |   |   |   |   |   |   |   |   |   mass < 30.6
|   |   |   |   |   |   |   |   |   |   |   plas < 124.5 : tested_positive (7/0) [7/3]
|   |   |   |   |   |   |   |   |   |   |   plas >= 124.5
|   |   |   |   |   |   |   |   |   |   |   |   preg < 4.5 : tested_positive (2/0) [1/0]
|   |   |   |   |   |   |   |   |   |   |   |   preg >= 4.5 : tested_negative (3/1) [1/0]
|   |   |   |   |   |   |   |   |   |   mass >= 30.6
|   |   |   |   |   |   |   |   |   |   |   age < 30.5 : tested_negative (10/0) [3/0]
|   |   |   |   |   |   |   |   |   |   |   age >= 30.5 : tested_positive (27/12) [19/9]
|   |   |   |   |   |   |   |   |   pedi >= 0.49 : tested_negative (4/0) [1/0]
|   |   |   |   |   |   |   |   pres >= 85 : tested_negative (11/0) [7/2]
|   |   |   |   |   |   |   mass >= 43.1 : tested_positive (5/0) [1/1]
|   |   |   |   |   age >= 53.5 : tested_negative (8/0) [3/1]
|   |   |   pedi >= 0.53
|   |   |   |   age < 26.5 : tested_negative (26/8) [8/2]
|   |   |   |   age >= 26.5
|   |   |   |   |   insu < 116
|   |   |   |   |   |   pedi < 0.67 : tested_positive (6/0) [6/2]
|   |   |   |   |   |   pedi >= 0.67
|   |   |   |   |   |   |   pedi < 0.73 : tested_negative (4/0) [3/0]
|   |   |   |   |   |   |   pedi >= 0.73 : tested_positive (14/5) [8/2]
|   |   |   |   |   insu >= 116 : tested_positive (15/0) [12/8]
plas >= 139.5
|   plas < 166.5
|   |   preg < 6.5
|   |   |   pedi < 0.33 : tested_negative (22/5) [10/5]
|   |   |   pedi >= 0.33
|   |   |   |   preg < 5.5
|   |   |   |   |   insu < 422.5 : tested_positive (28/8) [13/6]
|   |   |   |   |   insu >= 422.5 : tested_negative (2/0) [1/0]
|   |   |   |   preg >= 5.5 : tested_negative (5/1) [0/0]
|   |   preg >= 6.5
|   |   |   mass < 29 : tested_negative (6/3) [1/0]
|   |   |   mass >= 29 : tested_positive (21/0) [9/4]
|   plas >= 166.5 : tested_positive (52/6) [27/5]

Size of the tree : 49

Time taken to build model: 0.09 seconds

=== Evaluation on training set ===

Time taken to test model on training data: 0.14 seconds

=== Summary ===

Correctly Classified Instances         638               83.0729 %
Incorrectly Classified Instances       130               16.9271 %
Kappa statistic                          0.6313
Mean absolute error                      0.2498
Root mean squared error                  0.3534
Relative absolute error                 54.97   %
Root relative squared error             74.1514 %
Total Number of Instances              768     

=== Detailed Accuracy By Class ===

                 TP Rate  FP Rate  Precision  Recall   F-Measure  MCC      ROC Area  PRC Area  Class
                 0,858    0,220    0,879      0,858    0,868      0,632    0,885     0,922     tested_negative
                 0,780    0,142    0,746      0,780    0,763      0,632    0,885     0,778     tested_positive
Weighted Avg.    0,831    0,193    0,833      0,831    0,832      0,632    0,885     0,872     

=== Confusion Matrix ===

   a   b   <-- classified as
 429  71 |   a = tested_negative
  59 209 |   b = tested_positive