Merge pull request #509 from wcmc-its/development

Added publication types and remove JSON web tokens

pom.xml

@@ -282,7 +282,7 @@
 	    <dependency>
 	    	<groupId>edu.cornell.weill.reciter</groupId>
 	    	<artifactId>reciter-pubmed-model</artifactId>
-	    	<version>2.0.2</version>
+	    	<version>2.0.3</version>
 	    	<exclusions>
                 <exclusion>
                     <groupId>org.slf4j</groupId>
@@ -298,7 +298,7 @@
 	    <dependency>
 	    	<groupId>edu.cornell.weill.reciter</groupId>
 	    	<artifactId>reciter-article-model</artifactId>
-	    	<version>2.0.16</version>
+	    	<version>2.0.18</version>
 	    </dependency>
 	    <dependency>
 	    	<groupId>edu.cornell.weill.reciter</groupId>

src/main/java/reciter/algorithm/util/ArticleTranslator.java

@@ -1,3 +1,5 @@
+
+
 /*******************************************************************************
  * Licensed to the Apache Software Foundation (ASF) under one
  * or more contributor license agreements.  See the NOTICE file
@@ -162,6 +164,7 @@ public static ReCiterArticle translate(PubMedArticle pubmedArticle, ScopusArticl
                     ReCiterAuthor reCiterAuthor = new ReCiterAuthor(authorName, affiliation);
                     reCiterAuthor.setRank(i++);
                     reCiterAuthor.setOrcid(author.getOrcid());
+                    reCiterAuthor.setEqualContrib(author.getEqualContrib());
                     reCiterCoAuthors.addAuthor(reCiterAuthor);
                 }
             }
@@ -445,6 +448,10 @@ private static void determinePublicationTypeCanonical(ReCiterArticle reCiterArti
                 publicationTypeCanonical = "Letter";
             } else if(reCiterArticle.getPublicationTypePubmed().contains("Comment")) {
                 publicationTypeCanonical = "Comment";
+            } else if(reCiterArticle.getPublicationTypePubmed().contains("Preprint")) {
+                  publicationTypeCanonical = "Preprint";
+            } else if(reCiterArticle.getPublicationTypePubmed().contains("Published Erratum")) {
+                  publicationTypeCanonical = "Erratum";                
             } else if(reCiterArticle.getPublicationTypePubmed().contains("Consensus Development Conference")
                     ||
                     reCiterArticle.getPublicationTypePubmed().contains("Consensus Development Conference, NIH")
@@ -501,6 +508,8 @@ private static void determinePublicationTypeCanonical(ReCiterArticle reCiterArti
                     ||
                     reCiterArticle.getPublicationTypePubmed().contains("Comparative Study")
                     ||
+					reCiterArticle.getPublicationTypePubmed().contains("Clinical Trial")                    
+            		||
                     reCiterArticle.getPublicationTypePubmed().contains("Technical Report")) {
                 publicationTypeCanonical = "Academic Article";
             } else {

src/main/java/reciter/engine/ReCiterFeatureGenerator.java

@@ -119,11 +119,11 @@ public ReCiterFeature computeFeatures(UseGoldStandard mode,
 
         reCiterFeature.setCountSuggestedArticles(selectedArticles.size());
 
-        //Count of pending publications
+        // Count of pending publications
         List<ReCiterArticle> pendingArticles = selectedArticles
-        			.stream()
-        			.filter(reCiterArticle -> reCiterArticle.getTotalArticleScoreStandardized() >= filterScore && reCiterArticle.getGoldStandard() == 0)
-                    .collect(Collectors.toList());
+                .stream()
+                .filter(reCiterArticle -> reCiterArticle.getTotalArticleScoreStandardized() >= filterScore && reCiterArticle.getGoldStandard() == 0)
+                .collect(Collectors.toList());
         reCiterFeature.setCountPendingArticles(pendingArticles.size());
 
         List<Long> filteredArticles = selectedArticles.stream().map(article -> article.getArticleId()).collect(Collectors.toList());
@@ -133,20 +133,21 @@ public ReCiterFeature computeFeatures(UseGoldStandard mode,
         log.info("Analysis for uid=[" + identity.getUid() + "]");
         log.info("Precision=" + analysis.getPrecision());
         log.info("Recall=" + analysis.getRecall());
-
-	double accuracy = (analysis.getPrecision() + analysis.getRecall()) / 2.0;
-
+
+        // Calculate accuracy using the new formula
+        double accuracy = (double)(analysis.getTruePositiveList().size() + analysis.getTrueNegativeList().size()) / (double)(analysis.getTruePositiveList().size() + analysis.getTrueNegativeList().size() + analysis.getFalsePositiveList().size() + analysis.getFalseNegativeList().size());
+
         log.info("Accuracy=" + accuracy);
-
+        
         log.info("True Positive List [" + analysis.getTruePositiveList().size() + "]: " + analysis.getTruePositiveList());
         log.info("True Negative List: [" + analysis.getTrueNegativeList().size() + "]: " + analysis.getTrueNegativeList());
         log.info("False Positive List: [" + analysis.getFalsePositiveList().size() + "]: " + analysis.getFalsePositiveList());
         log.info("False Negative List: [" + analysis.getFalseNegativeList().size() + "]: " + analysis.getFalseNegativeList());
         log.info("\n");
 
-        // overall accuracy
-        reCiterFeature.setOverallAccuracy((analysis.getPrecision() + analysis.getRecall()) / 2);
-	    
+        // Set overall accuracy using the new formula
+        reCiterFeature.setOverallAccuracy(accuracy);
+
         // precision
         reCiterFeature.setPrecision(analysis.getPrecision());
 
@@ -364,6 +365,11 @@ else if (reCiterArticle.getGoldStandard() == -1) {
                         reCiterArticleAuthorFeature.setOrcid(matcher.group());
                     }
                 }
+                //EqualContrib
+                if(reCiterArticleAuthor.getEqualContrib()!=null && !reCiterArticleAuthor.getEqualContrib().isEmpty())
+                {
+                	reCiterArticleAuthorFeature.setEqualContrib(reCiterArticleAuthor.getEqualContrib());
+                }
 
 
                 reCiterArticleAuthorFeatures.add(reCiterArticleAuthorFeature);