Highlight fixes

Author: JoanGi

examples/evaluation/Gender.datadesc

@@ -88,11 +88,11 @@ Dataset: Gender
                             ofType: Categorical 
 
                 Statistics:
-                    Categorical Distribution:
+                    Categoric Distribution:
                         ["": 0%]
                     Quality Metrics: 
                         Sparsity: 8.9 
-                        Noisy labels "Please considers some humans errors in the annotation process could be done"
+                        Noisy labels: "Please considers some humans errors in the annotation process could be done"
                 Is sample: 
                     "It is a sample of all possible documents. It is not intended to be representative 
                     (in fact, it is known to be quite non-representative): it was specifically designed 

examples/evaluation/Melanoma.datadesc

@@ -123,7 +123,7 @@ Dataset: Melanoma_Classification_Dataset
                         ofType: Categorical
                         Statistics:
                                 Mode: "Famela"
-                                Categorical Distribution:
+                                Categoric Distribution:
                                     [
                                         "Male": 40%,
                                         "Famela": 55%,
@@ -161,9 +161,9 @@ Dataset: Melanoma_Classification_Dataset
                         count: 33126
                         ofType: Categorical
                         Statistics:
-                                Categorical Distribution:
+                                Categoric Distribution:
                                     [
-                                        "beningnant": 80%,
+                                     "beningnant": 80%,
                                      "malignant": 20%
                                     ]
         Statistics: 
@@ -208,7 +208,6 @@ Dataset: Melanoma_Classification_Dataset
                         (https://www.kaggle.com/c/siim- isic-melanoma- classification/discussion/161943)."
 
                 Related Instances: skinImages
-                Social Issues: PrivacyIssue1
                 How data is collected: Manual Human Curator
                 When data was collected: 
                     "Images were originally collected by imaging centers during 1998 - 2019; this dataset was curated from those image 
@@ -260,7 +259,6 @@ Dataset: Melanoma_Classification_Dataset
                     Who collects the data: "Internal Medical staff"
                     Type Internal
                     Country/Region: "Australia"
-                Social Issues: raceRepresentative
                 Label Requirements: 
                     Requirement: "1) Images containing any potentially identifying features, such as jewelry or tattoos, or from patients without 
                         at least three qualifying images were excluded during quality assurance review."
@@ -303,12 +301,13 @@ Dataset: Melanoma_Classification_Dataset
 
 
     Social Concerns: 
-      Rationale: 'Rationale'
+      Rationale: 'Dataset may not be representative of the real world data, and the cavenience sample is not representative of general incidence of melanoma'
       Social Issue: raceRepresentative
         IssueType: Bias
-        Related Attributes: ImageId
+        Related Attributes: 
+            attribute: ImageId  
         Description: "Dataset is not representative with respect to darker skin types"
       Social Issue: generalIncidence
-        IssueType: Social Impact
+        IssueType: Social Impact        
         Description: "Dataset is a convenience sample and is not representative of general incidence of melanoma"
 

examples/evaluation/Polarity.datadesc

@@ -5,10 +5,11 @@ Dataset: Polarity
         Version: v0001
         Release Date: 13-12-2010
         Citation: 
-            Raw Citation:
-                "Bo Pang and Lillian Lee. 2004. A sentimental education: Sentiment analysis using
-                subjectivity summarization based on minimum cuts. In Proceedings of the 42nd Annual
-                Meeting of the Association for Computational Linguistics. 271."
+            doi: "10.5281/zenodo.569"
+            authors: "Bo Pang and Lillian Lee"
+            title: "A sentimental education: Sentiment analysis using subjectivity summarization based on minimum cuts"
+            publisher: "Springer, proceedings of the 42nd Annual Meeting of the Association for Computational Linguistics"
+            year: "2004"
         Description:  
             Purposes: 
                 "The dataset was created to enable research on predicting sentiment polarity—i.e., given a piece of English text, predict whether
@@ -116,7 +117,7 @@ Dataset: Polarity
                           description: "The label annotated by the reviewrs " 
                           ofType: Categorical
                           Statistics:
-                            Categorical Distribution:
+                            Categoric Distribution:
                             [
                              "pos": 50%,
                              "neg": 50%
@@ -126,7 +127,7 @@ Dataset: Polarity
                         Quality Metrics:
                             Sparsity: 00
                             Completeness: 100
-                            Class Balance "attribute 'tag': 50% positive, 50% negative"
+                            Class Balance: "attribute 'tag': 50% positive, 50% negative"
 
                 Dependencies: 
                     Description: "The dataset is entirely self-contained."
@@ -191,7 +192,7 @@ Dataset: Polarity
             Label: Tag
             Description: "The label is the positive/negative sentiment polarity rating derived
             from the star rating"
-        Mapping: tag
+            Mapping: ImageId
         Label Requirements:
           Requirement:                     
             "- In order to obtain more accurate rating decisions, the maximum