Hello,
I tried to perform a data classification in WEKA through the J48 algorithm. There was no problem with the train set and cross-validation, but when I tried "supplied test set" this error appeared and the result was showed as a question mark. The results page is attached. How can I fix this problem?
I managed to do the prediction but it still shows inputmappedcalssifier. I know that are few reasons for that I just don't know how to get rid of it.
Diana Iagar,
InputMapped is used, in your case, to fix the problem of data incompatibility. Your training and test set are having different structures. I have fixed your data (so InputMapped is not going to be used) and attached them here. Now, if you need to run J48, it will not work as your data have string attributes; hence, you need to wrap J48 in FilteredClassifier and use StringToWordVector as a filter parameter. Once, you do so, you will get the following result after loading the test set data via the Supplied test set option.
Note that you will still get "?" in the statistical summary which is not an error here. It is just that you don't initially have labels in your test set data to compare the predicted labels with. I have explained this to you previously here: https://www.researchgate.net/post/How-to-fix-this-error-in-weka-train-and-test-set-are-not-compatible
HTH.
Dr. Samer Sarsam
=====================
=== Predictions on test set ===
inst# actual predicted error prediction
1 1:? 1:bad 1
2 1:? 2:good 1
3 1:? 1:bad 1
4 1:? 1:bad 1
5 1:? 2:good 1
6 1:? 2:good 1
7 1:? 2:good 1
8 1:? 2:good 1
9 1:? 1:bad 1
10 1:? 1:bad 1
11 1:? 2:good 1
12 1:? 2:good 1
13 1:? 2:good 1
14 1:? 1:bad 1
15 1:? 1:bad 1
16 1:? 1:bad 1
17 1:? 1:bad 1
18 1:? 1:bad 1
19 1:? 2:good 1
20 1:? 1:bad 1
21 1:? 2:good 1
22 1:? 2:good 1
23 1:? 1:bad 1
24 1:? 2:good 1
25 1:? 1:bad 1
26 1:? 1:bad 1
27 1:? 2:good 1
28 1:? 1:bad 1
29 1:? 1:bad 1
30 1:? 1:bad 1
31 1:? 1:bad 1
32 1:? 1:bad 1
33 1:? 2:good 1
34 1:? 1:bad 1
35 1:? 1:bad 1
36 1:? 1:bad 1
37 1:? 1:bad 1
38 1:? 1:bad 1
39 1:? 2:good 1
40 1:? 2:good 1
41 1:? 2:good 1
42 1:? 1:bad 1
43 1:? 1:bad 1
44 1:? 1:bad 1
45 1:? 1:bad 1
46 1:? 1:bad 1
47 1:? 1:bad 1
48 1:? 2:good 1
49 1:? 2:good 1
50 1:? 1:bad 1
51 1:? 1:bad 1
52 1:? 2:good 1
53 1:? 1:bad 1
54 1:? 1:bad 1
55 1:? 1:bad 1
56 1:? 1:bad 1
57 1:? 2:good 1
58 1:? 2:good 1
59 1:? 1:bad 1
60 1:? 2:good 1
61 1:? 1:bad 1
62 1:? 2:good 1
63 1:? 2:good 1
64 1:? 1:bad 1
65 1:? 1:bad 1
66 1:? 1:bad 1
67 1:? 2:good 1
68 1:? 2:good 1
69 1:? 2:good 1
70 1:? 2:good 1
71 1:? 1:bad 1
72 1:? 1:bad 1
73 1:? 1:bad 1
74 1:? 1:bad 1
75 1:? 2:good 1
76 1:? 2:good 1
77 1:? 1:bad 1
78 1:? 1:bad 1
79 1:? 1:bad 1
80 1:? 2:good 1
81 1:? 2:good 1
82 1:? 2:good 1
83 1:? 2:good 1
84 1:? 2:good 1
85 1:? 1:bad 1
86 1:? 1:bad 1
87 1:? 2:good 1
88 1:? 1:bad 1
89 1:? 1:bad 1
90 1:? 2:good 1
91 1:? 1:bad 1
92 1:? 1:bad 1
93 1:? 1:bad 1
94 1:? 2:good 1
95 1:? 1:bad 1
96 1:? 1:bad 1
97 1:? 2:good 1
98 1:? 1:bad 1
99 1:? 1:bad 1
100 1:? 2:good 1
101 1:? 1:bad 1
102 1:? 2:good 1
103 1:? 1:bad 1
104 1:? 2:good 1
105 1:? 1:bad 1
106 1:? 1:bad 1
107 1:? 2:good 1
108 1:? 2:good 1
109 1:? 1:bad 1
110 1:? 2:good 1
111 1:? 2:good 1
112 1:? 1:bad 1
113 1:? 1:bad 1
114 1:? 1:bad 1
115 1:? 1:bad 1
116 1:? 2:good 1
117 1:? 2:good 1
118 1:? 2:good 1
119 1:? 1:bad 1
120 1:? 1:bad 1
121 1:? 1:bad 1
122 1:? 1:bad 1
123 1:? 2:good 1
124 1:? 2:good 1
125 1:? 2:good 1
126 1:? 1:bad 1
127 1:? 1:bad 1
128 1:? 2:good 1
129 1:? 2:good 1
130 1:? 1:bad 1
131 1:? 1:bad 1
132 1:? 2:good 1
133 1:? 1:bad 1
134 1:? 1:bad 1
135 1:? 2:good 1
136 1:? 1:bad 1
137 1:? 1:bad 1
138 1:? 1:bad 1
139 1:? 2:good 1
140 1:? 1:bad 1
141 1:? 2:good 1
142 1:? 2:good 1
143 1:? 1:bad 1
144 1:? 1:bad 1
145 1:? 2:good 1
146 1:? 1:bad 1
=== Evaluation on test set ===
Time taken to test model on supplied test set: 0.14 seconds
=== Summary ===
Total Number of Instances 0
Ignored Class Unknown Instances 146
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
? ? ? ? ? ? ? ? bad
? ? ? ? ? ? ? ? good
Weighted Avg. ? ? ? ? ? ? ? ?
=== Confusion Matrix ===
a b
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification