Thank you Shareef

The supplied test data should have a class attribute with ? character to predict which values fall between false or true

i understood so took back my comment

If you believe that getting question marks means there is an error, then I would like to tell you that it is normal in your case. There is no error, but the result you had is mainly generated from the *missing* of the actual class in your test set (supplieddataset.csv). In your scenario, we are more interested to see the predicted class by the algorithm (e.g., NB); hence, you need to select the "PlainText" option. After evaluating NB under 10-fold cross-validation and applying it to your test set, I had the below result.

HTH.

Dr. Samer Sarsam

==============================

=== Predictions on user test set ===

inst# actual predicted error prediction

1 1:? 1:FALSE 0.524

2 1:? 1:FALSE 0.559

3 1:? 1:FALSE 0.534

4 1:? 1:FALSE 0.57

5 1:? 1:FALSE 0.679

6 1:? 1:FALSE 0.619

7 1:? 1:FALSE 0.544

8 1:? 2:True 0.553

9 1:? 2:True 0.573

10 1:? 1:FALSE 0.518

11 1:? 1:FALSE 0.545

12 1:? 1:FALSE 0.516

13 1:? 2:True 0.525

14 1:? 2:True 0.553

15 1:? 1:FALSE 0.506

16 1:? 2:True 0.576

17 1:? 2:True 0.548

18 1:? 2:True 0.506

19 1:? 2:True 0.565

20 1:? 2:True 0.57

21 1:? 2:True 0.567

22 1:? 2:True 0.562

23 1:? 2:True 0.55

24 1:? 2:True 0.534

25 1:? 1:FALSE 0.522

26 1:? 1:FALSE 0.569

27 1:? 1:FALSE 0.535

28 1:? 1:FALSE 0.614

29 1:? 1:FALSE 0.598

30 1:? 1:FALSE 0.546

31 1:? 1:FALSE 0.559

32 1:? 1:FALSE 0.656

33 1:? 1:FALSE 0.526

34 1:? 1:FALSE 0.598

35 1:? 1:FALSE 0.646

36 1:? 1:FALSE 0.624

37 1:? 1:FALSE 0.564

38 1:? 1:FALSE 0.539

39 1:? 1:FALSE 0.685

40 1:? 1:FALSE 0.598

41 1:? 1:FALSE 0.602

42 1:? 1:FALSE 0.582

43 1:? 1:FALSE 0.577

44 1:? 1:FALSE 0.561

45 1:? 1:FALSE 0.596

46 1:? 1:FALSE 0.6

47 1:? 1:FALSE 0.619

48 1:? 1:FALSE 0.614

49 1:? 1:FALSE 0.597

50 1:? 1:FALSE 0.586

51 1:? 1:FALSE 0.612

52 1:? 1:FALSE 0.682

53 1:? 1:FALSE 0.713

54 1:? 1:FALSE 0.699

55 1:? 1:FALSE 0.711

56 1:? 1:FALSE 0.675

57 1:? 1:FALSE 0.646

58 1:? 1:FALSE 0.687

59 1:? 1:FALSE 0.642

60 1:? 1:FALSE 0.698

61 1:? 1:FALSE 0.692

62 1:? 1:FALSE 0.69

63 1:? 1:FALSE 0.684

64 1:? 1:FALSE 0.659

65 1:? 1:FALSE 0.729

66 1:? 1:FALSE 0.738

67 1:? 1:FALSE 0.704

68 1:? 1:FALSE 0.666

69 1:? 1:FALSE 0.651

70 1:? 1:FALSE 0.639

71 1:? 1:FALSE 0.611

72 1:? 1:FALSE 0.637

73 1:? 1:FALSE 0.618

74 1:? 1:FALSE 0.612

75 1:? 1:FALSE 0.66

76 1:? 1:FALSE 0.615

77 1:? 1:FALSE 0.625

78 1:? 1:FALSE 0.628

79 1:? 1:FALSE 0.609

80 1:? 1:FALSE 0.651

81 1:? 1:FALSE 0.645

82 1:? 1:FALSE 0.637

83 1:? 1:FALSE 0.623

84 1:? 1:FALSE 0.623

85 1:? 1:FALSE 0.637

86 1:? 1:FALSE 0.64

87 1:? 1:FALSE 0.598

88 1:? 1:FALSE 0.613

89 1:? 1:FALSE 0.63

90 1:? 1:FALSE 0.636

91 1:? 1:FALSE 0.625

92 1:? 1:FALSE 0.625

93 1:? 1:FALSE 0.625

94 1:? 1:FALSE 0.666

95 1:? 1:FALSE 0.636

96 1:? 1:FALSE 0.636

97 1:? 1:FALSE 0.641

98 1:? 1:FALSE 0.656

99 1:? 1:FALSE 0.653

100 1:? 1:FALSE 0.644

101 1:? 1:FALSE 0.641

102 1:? 1:FALSE 0.524

103 1:? 1:FALSE 0.587

104 1:? 1:FALSE 0.71

105 1:? 1:FALSE 0.652

106 1:? 1:FALSE 0.622

107 1:? 1:FALSE 0.67

108 1:? 1:FALSE 0.661

109 1:? 1:FALSE 0.679

110 1:? 1:FALSE 0.711

111 1:? 1:FALSE 0.687

112 1:? 1:FALSE 0.687

113 1:? 1:FALSE 0.673

114 1:? 1:FALSE 0.724

115 1:? 1:FALSE 0.602

116 1:? 1:FALSE 0.605

117 1:? 1:FALSE 0.624

118 1:? 1:FALSE 0.681

119 1:? 1:FALSE 0.625

120 1:? 1:FALSE 0.674

121 1:? 1:FALSE 0.646

122 1:? 1:FALSE 0.64

123 1:? 1:FALSE 0.597

124 1:? 1:FALSE 0.59

125 1:? 1:FALSE 0.587

126 1:? 1:FALSE 0.62

127 1:? 1:FALSE 0.591

128 1:? 1:FALSE 0.599

129 1:? 1:FALSE 0.56

130 1:? 2:True 0.508

131 1:? 1:FALSE 0.513

132 1:? 1:FALSE 0.506

133 1:? 2:True 0.539

134 1:? 2:True 0.529

135 1:? 2:True 0.523

136 1:? 1:FALSE 0.553

137 1:? 1:FALSE 0.6

138 1:? 1:FALSE 0.552

139 1:? 1:FALSE 0.607

140 1:? 1:FALSE 0.601

141 1:? 1:FALSE 0.605

142 1:? 1:FALSE 0.628

143 1:? 1:FALSE 0.643

144 1:? 1:FALSE 0.628

145 1:? 1:FALSE 0.527

146 1:? 1:FALSE 0.515

147 1:? 1:FALSE 0.626

148 1:? 1:FALSE 0.652

149 1:? 1:FALSE 0.624

150 1:? 1:FALSE 0.636

151 1:? 1:FALSE 0.552

152 1:? 2:True 0.503

153 1:? 1:FALSE 0.575

154 1:? 1:FALSE 0.576

155 1:? 1:FALSE 0.582

156 1:? 1:FALSE 0.6

157 1:? 1:FALSE 0.652

158 1:? 1:FALSE 0.655

159 1:? 1:FALSE 0.575

160 1:? 1:FALSE 0.561

161 1:? 1:FALSE 0.576

162 1:? 1:FALSE 0.564

163 1:? 1:FALSE 0.568

164 1:? 2:True 0.51

165 1:? 2:True 0.521

166 1:? 2:True 0.57

167 1:? 2:True 0.585

168 1:? 2:True 0.573

169 1:? 2:True 0.537

170 1:? 1:FALSE 0.531

171 1:? 1:FALSE 0.5

172 1:? 2:True 0.505

173 1:? 2:True 0.584

174 1:? 2:True 0.608

175 1:? 1:FALSE 0.555

176 1:? 1:FALSE 0.611

177 1:? 1:FALSE 0.609

178 1:? 1:FALSE 0.577

179 1:? 1:FALSE 0.581

180 1:? 1:FALSE 0.586

181 1:? 1:FALSE 0.542

182 1:? 1:FALSE 0.551

183 1:? 1:FALSE 0.548

184 1:? 1:FALSE 0.586

185 1:? 2:True 0.552

186 1:? 1:FALSE 0.556

187 1:? 1:FALSE 0.525

188 1:? 1:FALSE 0.518

189 1:? 1:FALSE 0.51

190 1:? 1:FALSE 0.54

191 1:? 1:FALSE 0.542

192 1:? 2:True 0.513

193 1:? 2:True 0.513

194 1:? 2:True 0.573

195 1:? 2:True 0.578

196 1:? 2:True 0.568

197 1:? 2:True 0.602

198 1:? 2:True 0.556

199 1:? 2:True 0.566

200 1:? 2:True 0.571

201 1:? 2:True 0.555

202 1:? 2:True 0.555

203 1:? 2:True 0.518

204 1:? 2:True 0.535

205 1:? 2:True 0.538

206 1:? 1:FALSE 0.519

207 1:? 2:True 0.518

208 1:? 2:True 0.505

209 1:? 2:True 0.51

210 1:? 2:True 0.504

211 1:? 2:True 0.619

212 1:? 2:True 0.512

213 1:? 2:True 0.526

214 1:? 2:True 0.519

215 1:? 1:FALSE 0.542

216 1:? 1:FALSE 0.542

217 1:? 1:FALSE 0.614

218 1:? 1:FALSE 0.568

219 1:? 1:FALSE 0.591

220 1:? 1:FALSE 0.591

221 1:? 1:FALSE 0.545

222 1:? 1:FALSE 0.59

223 1:? 1:FALSE 0.587

224 1:? 1:FALSE 0.524

225 1:? 1:FALSE 0.519

226 1:? 2:True 0.564

227 1:? 1:FALSE 0.515

228 1:? 1:FALSE 0.519

229 1:? 1:FALSE 0.53

230 1:? 1:FALSE 0.539

231 1:? 1:FALSE 0.557

232 1:? 1:FALSE 0.555

233 1:? 1:FALSE 0.568

234 1:? 2:True 0.552

235 1:? 2:True 0.591

236 1:? 2:True 0.543

237 1:? 2:True 0.554

238 1:? 2:True 0.566

239 1:? 1:FALSE 0.517

240 1:? 1:FALSE 0.507

241 1:? 1:FALSE 0.571

242 1:? 1:FALSE 0.542

243 1:? 1:FALSE 0.588

244 1:? 1:FALSE 0.557

245 1:? 1:FALSE 0.609

246 1:? 1:FALSE 0.597

247 1:? 1:FALSE 0.548

248 1:? 1:FALSE 0.558

249 1:? 1:FALSE 0.516

250 1:? 1:FALSE 0.558

251 1:? 1:FALSE 0.562

252 1:? 1:FALSE 0.547

253 1:? 1:FALSE 0.547

254 1:? 1:FALSE 0.536

255 1:? 2:True 0.538

256 1:? 2:True 0.51

257 1:? 2:True 0.6

258 1:? 2:True 0.59

259 1:? 2:True 0.574

260 1:? 2:True 0.523

261 1:? 2:True 0.579

262 1:? 2:True 0.501

263 1:? 2:True 0.573

264 1:? 2:True 0.57

265 1:? 2:True 0.567

266 1:? 2:True 0.572

267 1:? 2:True 0.553

268 1:? 2:True 0.503

269 1:? 2:True 0.52

270 1:? 2:True 0.535

271 1:? 2:True 0.576

272 1:? 2:True 0.519

273 1:? 2:True 0.532

274 1:? 2:True 0.528

275 1:? 2:True 0.548

276 1:? 1:FALSE 0.502

277 1:? 2:True 0.519

278 1:? 1:FALSE 0.51

279 1:? 1:FALSE 0.51

280 1:? 2:True 0.555

281 1:? 1:FALSE 0.521

282 1:? 2:True 0.532

283 1:? 2:True 0.528

284 1:? 2:True 0.508

285 1:? 2:True 0.56

286 1:? 2:True 0.512

287 1:? 1:FALSE 0.51

288 1:? 2:True 0.52

289 1:? 2:True 0.517

290 1:? 2:True 0.563

291 1:? 2:True 0.538

292 1:? 2:True 0.576

293 1:? 2:True 0.583

294 1:? 2:True 0.544

295 1:? 2:True 0.536

296 1:? 2:True 0.555

297 1:? 1:FALSE 0.501

298 1:? 1:FALSE 0.516

Dr Samer

With the assumption to calculate precision, recall,LOC area and accuracy of the NB algorithm.How can that be achieved?. Additionally, I'm testing algorithm accuracy with relation to what values are incorrectly or correctly classified

Daniel Chege

If you need to compute the classifier's accuracy, recall, ROC, etc, then simply invoke your classifier, e.g., NB, on the training set data (trainingdataset.csv, your data) and validate it via stratified 10-fold cross-validation (no need to have independent test set--supplieddataset.csv, your data). You can report the result of cross-validation which is fairly enough to assess your predictive model.

You can rely on the confusion matrix to see what values are incorrectly/correctly classified. WEKA also allows you visualizing this result as well via the "Visualize classifier errors" option that can be obtained from the result history.

Cheers,

Dr. Samer Sarsam

Thanks Dr Samer Sarsam.

The training data is randomly generated while as the test data is collected from my experiment. When I rely on the random training data for analysis; that assumption classifies my analysis as unreliable. in this case, it seems I'm not using the test data to predict the results.

From the Weka process you proposed before still the result outputs with ?. On the data I have identified the class attribute but still I'm getting the same results as before.