Front End - NLP - work in progress

src/main/java/org/texttechnologylab/project/gruppe_05_1/database/MongoPprUtils.java

@@ -8,6 +8,7 @@ import org.texttechnologylab.project.gruppe_05_1.database.domainimp.speeches.Spe
 import org.texttechnologylab.project.gruppe_05_1.domain.html.HtmlSpeech;
 import org.texttechnologylab.project.gruppe_05_1.domain.html.Parlamentarier;
 import org.texttechnologylab.project.gruppe_05_1.domain.html.ParlamentarierDetails;
+import org.texttechnologylab.project.gruppe_05_1.domain.nlp.*;
 import org.texttechnologylab.project.gruppe_05_1.domain.speaker.Membership;
 import org.texttechnologylab.project.gruppe_05_1.domain.speech.SpeechMetaData;
 import org.texttechnologylab.project.gruppe_05_1.util.GeneralUtils;
@@ -417,10 +418,19 @@ public class MongoPprUtils {
         }
     }
 
+    /**
+     * Liefert die Rede-Informationen für die Anzeige einer Rede:
+     * - die Rede-ID
+     * - Name und Fraktion des Redners
+     * - Die Inhalte der Rede
+     * @param key: Rede ID
+     * @return
+     */
     public static HtmlSpeech getSpeechByKey(String key) {
         Document filter = new Document("speechKey", key);
         Document speechDoc = getSpeechCollection().find(filter).first();
 
         return new HtmlSpeech(speechDoc);
     }
+
 }

src/main/java/org/texttechnologylab/project/gruppe_05_1/domain/html/HtmlSpeech.java

@@ -2,6 +2,8 @@ package org.texttechnologylab.project.gruppe_05_1.domain.html;
 
 import org.bson.Document;
 import org.texttechnologylab.project.gruppe_05_1.database.MongoDBHandler;
+import org.texttechnologylab.project.gruppe_05_1.domain.nlp.NlpInfo;
+import org.texttechnologylab.project.gruppe_05_1.domain.nlp.Topic;
 
 import java.util.ArrayList;
 import java.util.List;
@@ -13,6 +15,7 @@ public class HtmlSpeech {
     String                  speakerName;
     String                  fraction;
     List<SpeechContent> content = new ArrayList<>();
+    NlpInfo                 nlp = null;
 
     public HtmlSpeech() {
     }
@@ -30,6 +33,32 @@ public class HtmlSpeech {
                 addContent(new SpeechContent(contentDoc));
             }
         }
+
+        Document nlpDoc = (Document) doc.get("analysisResults");
+        nlp = readNlpInfo(nlpDoc);
+    }
+
+    private NlpInfo readNlpInfo(Document nlpDoc) {
+        if (nlpDoc == null) return null;
+        NlpInfo nlp = new NlpInfo();
+
+        // TODO: HERE
+        List<Document> tokensDocs = nlpDoc.get("tokens", MongoDBHandler.DOC_LIST_CLASS);
+
+        List<Document> sentencesDocs = nlpDoc.get("sentences", MongoDBHandler.DOC_LIST_CLASS);
+
+        List<Document> dependenciesDocs = nlpDoc.get("dependencies", MongoDBHandler.DOC_LIST_CLASS);
+
+        List<Document> namedEntitiesDocs = nlpDoc.get("namedEntities", MongoDBHandler.DOC_LIST_CLASS);
+
+        List<Document> sentimentsDocs = nlpDoc.get("sentiments", MongoDBHandler.DOC_LIST_CLASS);
+
+        List<Document> topicsDocs = nlpDoc.get("topics", MongoDBHandler.DOC_LIST_CLASS);
+        nlp.setTopics(Topic.readTopicsFromMongo(topicsDocs));
+
+        // TODO: Video
+
+        return nlp;
     }
 
     public String getSpeechKey() {
@@ -68,16 +97,26 @@ public class HtmlSpeech {
         content.add(contentLine);
     }
 
+    public NlpInfo getNlp() {
+        return nlp;
+    }
+
+    public void setNlp(NlpInfo nlp) {
+        this.nlp = nlp;
+    }
+
     @Override
     public boolean equals(Object o) {
         if (this == o) return true;
         if (!(o instanceof HtmlSpeech that)) return false;
-        return Objects.equals(speechKey, that.speechKey) && Objects.equals(speakerName, that.speakerName) && Objects.equals(fraction, that.fraction) && Objects.equals(content, that.content);
+        return Objects.equals(speechKey, that.speechKey) && Objects.equals(speakerName, that.speakerName)
+                && Objects.equals(fraction, that.fraction) && Objects.equals(content, that.content)
+                && Objects.equals(nlp, that.nlp);
     }
 
     @Override
     public int hashCode() {
-        return Objects.hash(speechKey, speakerName, fraction, content);
+        return Objects.hash(speechKey, speakerName, fraction, content, nlp);
     }
 
     @Override
@@ -87,6 +126,7 @@ public class HtmlSpeech {
                 .add("speakerName='" + speakerName + "'")
                 .add("fraction='" + fraction + "'")
                 .add("content=" + content)
+                .add("nlp=" + nlp)
                 .toString();
     }
 }

src/main/java/org/texttechnologylab/project/gruppe_05_1/domain/nlp/NlpInfo.java

@@ -1,7 +1,9 @@
 package org.texttechnologylab.project.gruppe_05_1.domain.nlp;
 
 import java.util.List;
+import java.util.Map;
 import java.util.Objects;
+import java.util.StringJoiner;
 
 public class NlpInfo {
     List<Token>         tokens;
@@ -10,7 +12,7 @@ public class NlpInfo {
     List<NamedEntity>   namedEntities;
     Sentiment           overallSentiment;       // Sentiment for the whole text ; kann null sein!
     List<Sentiment>     sentiments;             // sentiments for the respective sentences (eine Liste von 0..n Elementen)
-    List<Topic>         topic;
+    List<Topic>         topics;
     List<Pos>           posList;
 
     VideoInformation    videoInformation;
@@ -63,12 +65,12 @@ public class NlpInfo {
         this.sentiments = sentiments;
     }
 
-    public List<Topic> getTopic() {
-        return topic;
+    public List<Topic> getTopics() {
+        return topics;
     }
 
-    public void setTopic(List<Topic> topic) {
-        this.topic = topic;
+    public void setTopics(List<Topic> topics) {
+        this.topics = topics;
     }
 
     public List<Pos> getPosList() {
@@ -94,12 +96,28 @@ public class NlpInfo {
         return Objects.equals(tokens, nlpInfo.tokens) && Objects.equals(sentences, nlpInfo.sentences)
                 && Objects.equals(dependencies, nlpInfo.dependencies) && Objects.equals(namedEntities, nlpInfo.namedEntities)
                 && Objects.equals(overallSentiment, nlpInfo.overallSentiment) && Objects.equals(sentiments, nlpInfo.sentiments)
-                && Objects.equals(topic, nlpInfo.topic) && Objects.equals(posList, nlpInfo.posList)
+                && Objects.equals(topics, nlpInfo.topics) && Objects.equals(posList, nlpInfo.posList)
                 && Objects.equals(videoInformation, nlpInfo.videoInformation);
     }
 
     @Override
     public int hashCode() {
-        return Objects.hash(tokens, sentences, dependencies, namedEntities, overallSentiment, sentiments, topic, posList, videoInformation);
+        return Objects.hash(tokens, sentences, dependencies, namedEntities, overallSentiment, sentiments, topics, posList, videoInformation);
     }
+
+    @Override
+    public String toString() {
+        return new StringJoiner(", ", NlpInfo.class.getSimpleName() + "[", "]")
+                .add("tokens=" + tokens)
+                .add("sentences=" + sentences)
+                .add("dependencies=" + dependencies)
+                .add("namedEntities=" + namedEntities)
+                .add("overallSentiment=" + overallSentiment)
+                .add("sentiments=" + sentiments)
+                .add("topics=" + topics)
+                .add("posList=" + posList)
+                .add("videoInformation=" + videoInformation)
+                .toString();
+    }
+
 }

src/main/java/org/texttechnologylab/project/gruppe_05_1/domain/nlp/Topic.java

@@ -1,18 +1,19 @@
 package org.texttechnologylab.project.gruppe_05_1.domain.nlp;
 
-import java.util.Objects;
-import java.util.StringJoiner;
+import org.bson.Document;
+
+import java.util.*;
+import java.util.stream.Collectors;
 
 public class Topic {
     String      topic;
-    double      score;
-    // tags        TODO
+    Double      score;
     String      text;
 
     public Topic() {
     }
 
-    public Topic(String topic, double score, String text) {
+    public Topic(String topic, Double score, String text) {
         this.topic = topic;
         this.score = score;
         this.text = text;
@@ -26,11 +27,11 @@ public class Topic {
         this.topic = topic;
     }
 
-    public double getScore() {
+    public Double getScore() {
         return score;
     }
 
-    public void setScore(double score) {
+    public void setScore(Double score) {
         this.score = score;
     }
 
@@ -62,4 +63,45 @@ public class Topic {
                 .add("text='" + text + "'")
                 .toString();
     }
+
+
+    /**
+     * Die Topics-Dokumente (Speech --> analysisResults --> topics) aus der MongoDB lesen
+     * @param topicsDocs Eine Liste von Mongo-Dokumenten
+     * @return Eine Liste der Topics
+     */
+    public static List<Topic> readTopicsFromMongo(List<Document> topicsDocs) {
+        List<Topic> topics = new ArrayList<>();
+        for (Document doc : topicsDocs) {
+            topics.add(new Topic(doc.getString("topic"),
+                                 doc.getDouble("score"),
+                                 doc.getString("text")
+                    ));
+        }
+        return topics;
+    }
+
+
+    /**
+     * Topic-Informationen "verdichten":
+     * Ausgangssituation: eine Liste mit mehreren Topics. Ein Topic kann in dieser Liste mehrfach vorkommen.
+     * Man will wissen, welche Score hat jeden Topic. Hier werden die Werte der jeweiligen Topics summiert.
+     *
+     * @param topicsList
+     * @return Map<String, Double>
+     */
+    public static Map<String, Double> condenseTopicInformation(List<Topic> topicsList) {
+        Map<String, Double> condensedTopicInfo = new HashMap<>();
+
+        for (Topic t : topicsList) {
+            Double oldValue = condensedTopicInfo.get(t.getTopic());
+            if (oldValue != null) {
+                condensedTopicInfo.replace(t.getTopic(), oldValue + t.getScore());
+            } else {
+                condensedTopicInfo.put(t.getTopic(), t.getScore());
+            }
+        }
+
+        return condensedTopicInfo;
+    }
 }

src/main/java/org/texttechnologylab/project/gruppe_05_1/rest/SpeechController.java

@@ -5,12 +5,14 @@ import io.javalin.openapi.*;
 import org.texttechnologylab.project.gruppe_05_1.database.MongoPprUtils;
 import org.texttechnologylab.project.gruppe_05_1.domain.html.HtmlSpeech;
 import org.texttechnologylab.project.gruppe_05_1.domain.html.ParlamentarierDetails;
+import org.texttechnologylab.project.gruppe_05_1.domain.nlp.Topic;
 import org.texttechnologylab.project.gruppe_05_1.domain.speech.SpeechMetaData;
 import org.texttechnologylab.project.gruppe_05_1.xml.speeches.Interfaces.Speech;
 
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
+import java.util.stream.Collectors;
 
 public class SpeechController {
     /**
@@ -68,6 +70,15 @@ public class SpeechController {
         HtmlSpeech speech = MongoPprUtils.getSpeechByKey(redeId);
         attributes.put("s", speech);
 
+        // NLP: Topic
+        if ((speech.getNlp() != null) && (speech.getNlp().getTopics() != null)) {
+            Map<String, Double> topics = Topic.condenseTopicInformation(speech.getNlp().getTopics());   // Daten "verdichten"...
+            speech.getNlp().setTopics(
+                    topics.entrySet().stream()
+                            .map(me -> new Topic(me.getKey(), me.getValue(), null))
+                            .collect(Collectors.toList()));
+        }
+
         ctx.render("speech.ftl", attributes);
     }
 

src/main/java/org/texttechnologylab/project/gruppe_05_1/website/templates/nlp.ftl

@@ -0,0 +1,35 @@
+    <#if s.nlp.topics??>
+         <h3>Topics Information (als Bubble Chart)</h3>
+         <#assign condenseTopicInformation = s.nlp.topics>
+         <#include "topicsBubbleChart.ftl">
+    <#else>
+        <h3>Keine Topics Information für diese Rede verfügbar</h3>
+    </#if>
+
+
+    <#if s.nlp.posList??>
+         <h3>POS Information (als Bar Chart)</h3>
+         <#assign posList = s.nlp.posList>
+         <#include "posBarChart.ftl">
+    <#else>
+        <h3>Keine POS Information verfügbar für diese Rede verfügbar</h3>
+    </#if>
+
+
+    <#if s.nlp.sentiments??>
+         <h3>SentimentsInformation (als Radar Chart)</h3>
+         <#assign overallSentiment = s.nlp.overallSentiment>
+         <#assign sentiments = s.nlp.sentiments>
+         <#include "sentimentsRadarChart.ftl">
+    <#else>
+        <h3>Keine Sentiments Information für diese Rede verfügbar</h3>
+    </#if>
+
+
+    <#if s.nlp.namedEntities??>
+         <h3>Named Entities Information (als Sunburst Chart)</h3>
+         <#assign nea = s.nlp.namedEntities>
+         <#include "namedEntitiesSunburstChart.ftl">
+    <#else>
+        <h3>Keine Named Entities Information für diese Rede verfügbar</h3>
+    </#if>    

src/main/java/org/texttechnologylab/project/gruppe_05_1/website/templates/speech.ftl

@@ -22,11 +22,19 @@
 
     <h2>Rede ${s.speechKey} </h2>
 
-    <main>
-        <#list s.content as c>
-            <#include "speechContent.ftl">
-        </#list>
-    </main>
+    <#list s.content as c>
+        <#include "speechContent.ftl">
+    </#list>
+
+    <br><br>
+    <#if s.nlp??>
+         <h2>NLP Information</h2>
+         <#assign nlp = "${s.nlp}">
+         <#include "nlp.ftl">
+    <#else>
+        <h2>Keine NLP Information verfügbar für diese Rede</h2>
+    </#if>
+    <br> <br>
 
 
 </body>