Merge pull request #8 from gicraveiro/remove_empty_sentences_and_rerun

Remove empty sentences and rerun
+
Updated and cleaned AI classifier code file + deleted old files

.gitignore

@@ -1 +1,2 @@
-google_key.json
+google_key.json
+MEMORY "CPU" Sislab

features.txt

@@ -1,74 +1,76 @@
 privacy
-people
-system
-safe
-possible
-should
-experience
-safer
-right
 secure
 community
-believe
+should
+possible
 express
-we
-can
-use
-cookies
+believe
+right
+people
+your
 you
-data
+we
+they
+the
+safe
 or
-content
-with
+is
 information
-they
+data
+cookies
+no
+experience
+use
+protect
+with
+providers
+limited
 learn
-more
-provide
+if
 device
-understand
 days
-collect
-who
-the
-your
-is
-to be
+companies
+can
+automatically
+also
+about
 you should
-help keep
-provide you
-should be
-and protect
+to be
 when people
-to enable
-and privacy
-this right
+and protect
 public interest
-what you
+for people
 have a
+you share
+to enable
+this right
+provide you
+it with
+help keep
+are not
+advertising and
 you with
-for people
-use cookies
+with the
+you use
 you can
-20/09/2021 14
-more about
-help us
-the information
-information about
-we may
-and others
-products including
-not about
-these terms
-to build
 with facebook
-who you
-advertising and
-that help
+we may
 we collect
+we also
+the information
+the facebook
+that help
+such as
+share it
 share and
+products including
+post or
 not at
-services that
-share it
-it with
-are not
+more about
+information about
+help us
+cookies to
+and privacy
+and others
+ads to
+should be

output/AI Classifier/1labelPredictionsStatsDev.txt

@@ -1,14 +1,14 @@
-Performance measures - Unigram Dictionary - MLP
+Performance measures - Mixed Dictionary - Adaboost
 
 Dev set:
 
-Precision macro: 0.76
-Precision Individually: [0.65  1.    0.886 0.6   0.667]
-Recall macro: 0.665
-Recall Individually: [1.    0.5   0.886 0.273 0.667]
-F1 Score micro: 0.771
-F1 Score macro: 0.676
-F1 Score weighted: 0.753
-F1 Score Individually: [0.788 0.667 0.886 0.375 0.667]
+Precision macro: 0.425
+Precision Individually: [0.5   0.    0.696 0.5   0.429]
+Recall macro: 0.389
+Recall Individually: [0.154 0.    0.941 0.182 0.667]
+F1 Score micro: 0.609
+F1 Score macro: 0.365
+F1 Score weighted: 0.549
+F1 Score Individually: [0.235 0.    0.8   0.267 0.522]
 
 

output/AI Classifier/1labelPredictionsStatsTest.txt

@@ -1,14 +1,14 @@
-Performance measures - Unigram Dictionary - MLP Word Embeddings
+Performance measures - Mixed Dictionary - Adaboost
 
 Test set:
 
-Precision macro: 0.534
-Precision Individually: [0.778 0.    0.73  0.75  0.412]
-Recall macro: 0.469
-Recall Individually: [0.5   0.    0.871 0.273 0.7  ]
-F1 Score micro: 0.657
-F1 Score macro: 0.464
-F1 Score weighted: 0.638
-F1 Score Individually: [0.609 0.    0.794 0.4   0.519]
+Precision macro: 0.425
+Precision Individually: [0.8   0.    0.574 0.25  0.5  ]
+Recall macro: 0.315
+Recall Individually: [0.286 0.    1.    0.091 0.2  ]
+F1 Score micro: 0.567
+F1 Score macro: 0.314
+F1 Score weighted: 0.49
+F1 Score Individually: [0.421 0.    0.729 0.133 0.286]