Configuration File=file:/home/saggion/temp/ml-opinion/cross-validation/stars/root/config.xml ************************* A new session for NLP learning is starting. No feature list file in initialisation phrase. No label list file in initialisation phrase. No chunk length statistics list file in initialisation phrase. Pre-processing the 111 documents... Learning starts. For the information about this learning see the log file /home/saggion/temp/ml-opinion/cross-validation/stars/root/savedFiles/logFileForNLPLearning.save The number of threads used is 1 ** Evaluation mode: Kfold k=5, numDoc=111, len=22. *** Averaged results for each label over 5 runs as: Results of single label: 0 LabelName=1_Star_Review, number of instances=76 (correct, paritalCorrect, spurious, missing)= (12.8, 0.0, 3.0, 6.2); (precision, recall, F1)= (0.8225113, 0.6240529, 0.68918097); Lenient: (0.8225113, 0.6240529, 0.68918097) 1 LabelName=2_Star_Review, number of instances=23 (correct, paritalCorrect, spurious, missing)= (1.2, 0.0, 0.0, 4.6); (precision, recall, F1)= (0.6, 0.16904762, 0.25714284); Lenient: (0.6, 0.16904762, 0.25714284) 2 LabelName=3_Star_Review, number of instances=32 (correct, paritalCorrect, spurious, missing)= (0.8, 0.0, 0.2, 7.2); (precision, recall, F1)= (0.6, 0.11269841, 0.18603176); Lenient: (0.6, 0.11269841, 0.18603176) 3 LabelName=4_Star_Review, number of instances=193 (correct, paritalCorrect, spurious, missing)= (9.0, 0.0, 3.2, 38.6); (precision, recall, F1)= (0.7394017, 0.19508964, 0.30647406); Lenient: (0.7394017, 0.19508964, 0.30647406) 4 LabelName=5_Star_Review, number of instances=665 (correct, paritalCorrect, spurious, missing)= (154.0, 0.0, 41.8, 6.8); (precision, recall, F1)= (0.7855043, 0.95784295, 0.86262476); Lenient: (0.7855043, 0.95784295, 0.86262476) Overall results as: (correct, paritalCorrect, spurious, missing)= (177.8, 0.0, 48.2, 63.4); (precision, recall, F1)= (0.78726965, 0.73840994, 0.7620047); Lenient: (0.78726965, 0.73840994, 0.7620047) This learning session finished!. Pre-processing the 111 documents... Learning starts. For the information about this learning see the log file /home/saggion/temp/ml-opinion/cross-validation/stars/root/savedFiles/logFileForNLPLearning.save The number of threads used is 1 ** Training mode: time for NLP features: 5 time for fv: 1 time for filtering: 0 time for NLP training: 6 This learning session finished!. The 10 most significiant positive NLP feature for class: 5_Star_Review: 0: 0 _Sent2gram_very<> -- 2.4727945 1: 0 _Sent2gram_excellent<> -- 1.940732 2: 0 _Sent2gram_future<> -- 1.5263237 3: 0 _Sent2gram_experience<> -- 1.5178185 4: 0 _Sent2gram_always<> -- 1.396251 5: 0 _Sent2gram_yet<> -- 1.3091754 6: 0 _Sent2gram_buy<> -- 1.2263054 7: 0 _Sent2gram_great<> -- 1.2059165 8: 0 _Sent2gram_1<> -- 1.1904016 9: 0 _Sent2gram_two<> -- 1.1893711 The 10 most significiant positive NLP feature for class: 4_Star_Review: 0: 0 _Sent2gram_ok<> -- 2.1023943 1: 0 _Sent2gram_test<> -- 1.5787487 2: 0 _Sent2gram_befor<> -- 1.5635328 3: 0 _Sent2gram_think<> -- 1.4067355 4: 0 _Sent2gram_wasnt<> -- 1.3715109 5: 0 _Sent2gram_couple<> -- 1.3509235 6: 0 _Sent2gram_but<> -- 1.2988269 7: 0 _Sent2gram_however<> -- 1.2684104 8: 0 _Sent2gram_go<> -- 1.133596 9: 0 _Sent2gram_struggle<> -- 1.0533196 The 10 most significiant positive NLP feature for class: 3_Star_Review: 0: 0 _Sent2gram_another<> -- 1.1479185 1: 0 _Sent2gram_sent<> -- 0.97107625 2: 0 _Sent2gram_didnt<> -- 0.8722273 3: 0 _Sent2gram_improve<> -- 0.869917 4: 0 _Sent2gram_fine<> -- 0.8620054 5: 0 _Sent2gram_wrong<> -- 0.76554185 6: 0 _Sent2gram_-<> -- 0.74257696 7: 0 _Sent2gram_opposite<> -- 0.7080652 8: 0 _Sent2gram_satisfactory<> -- 0.68374425 9: 0 _Sent2gram_end<> -- 0.6717117 The 10 most significiant positive NLP feature for class: 1_Star_Review: 0: 0 _Sent2gram_they<> -- 1.501809 1: 0 _Sent2gram_/<> -- 1.3815038 2: 0 _Sent2gram_stay<> -- 1.3677018 3: 0 _Sent2gram_worst<> -- 1.3557749 4: 0 _Sent2gram_tell<> -- 1.2988111 5: 0 _Sent2gram_no<> -- 1.234583 6: 0 _Sent2gram_cancelled<> -- 1.2160219 7: 0 _Sent2gram_avoid<> -- 1.1711732 8: 0 _Sent2gram_!<> -- 1.1656644 9: 0 _Sent2gram_call<> -- 1.1033683 The 10 most significiant positive NLP feature for class: 2_Star_Review: 0: 0 _Sent2gram_shirt<> -- 1.0346017 1: 0 _Sent2gram_ball<> -- 0.9873586 2: 0 _Sent2gram_waited<> -- 0.88756883 3: 0 _Sent2gram_eg<> -- 0.6697718 4: 0 _Sent2gram_toshiba<> -- 0.64626956 5: 0 _Sent2gram_dvd<> -- 0.5873925 6: 0 _Sent2gram_11<> -- 0.56501675 7: 0 _Sent2gram_be<> -- 0.5394484 8: 0 _Sent2gram_log<> -- 0.5153032 9: 0 _Sent2gram_royal<> -- 0.5153032
