mathieu/CAU-NLP-Spring-2019-PQ-Man-Server
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Put the webserver

Browse Source
This commit is contained in:
Mathieu Sanchez 2019-06-12 12:05:50 +09:00
parent 1c36a2676a
commit cf33115501
5 changed files with 156 additions and 110 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
CAu_NLP_2019/pom.xml
View File

@ -35,6 +35,46 @@
<target>1.8</target> <target>1.8</target>
</configuration> </configuration>
</plugin> </plugin>
<plugin>
<artifactId>maven-assembly-plugin</artifactId>
<executions>
<execution>
<phase>package</phase>
<goals>
<goal>single</goal>
</goals>
</execution>
</executions>
<configuration>
<archive>
<manifest>
<mainClass>com.nlp.App</mainClass>
</manifest>
</archive>
<descriptorRefs>
<descriptorRef>jar-with-dependencies</descriptorRef>
</descriptorRefs>
</configuration>
</plugin>
<!-- Make this jar executable -->
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-jar-plugin</artifactId>
<configuration>
<!-- DO NOT include log4j.properties file in your Jar -->
<excludes>
<exclude>**/log4j.properties</exclude>
</excludes>
<archive>
<manifest>
<!-- Jar file entry point -->
<mainClass>com.nlp.App</mainClass>
</manifest>
</archive>
</configuration>
</plugin>
</plugins> </plugins>
</build> </build>
<groupId>PQMAN</groupId> <groupId>PQMAN</groupId>

4
CAu_NLP_2019/src/main/java/App.java → CAu_NLP_2019/src/main/java/com/nlp/App.java
View File

@ -1,3 +1,5 @@
package com.nlp;
import com.google.gson.Gson; import com.google.gson.Gson;
import org.jsoup.Jsoup; import org.jsoup.Jsoup;
import org.jsoup.nodes.Document; import org.jsoup.nodes.Document;
@ -28,8 +30,6 @@ public class App {
String[] url = {"https://www.thesun.co.uk/?s=", "https://www.bbc.co.uk/search?q=", "https://www.skysports.com/search?q="}; String[] url = {"https://www.thesun.co.uk/?s=", "https://www.bbc.co.uk/search?q=", "https://www.skysports.com/search?q="};
List<List<String>> documents = new ArrayList<>(); List<List<String>> documents = new ArrayList<>();
List<List<String>> result = new ArrayList<>();
List<String> result2 = new ArrayList<>();
String document; String document;
List<String> goodUrl = new ArrayList<>(); List<String> goodUrl = new ArrayList<>();
Article data = new Article(); Article data = new Article();

126
CAu_NLP_2019/src/main/java/Article.java → CAu_NLP_2019/src/main/java/com/nlp/Article.java
View File

@ -1,63 +1,65 @@
import java.util.ArrayList; package com.nlp;
public class Article { import java.util.ArrayList;
public Article() { public class Article {
date = new ArrayList<Integer>();
Headline = new ArrayList<String>(); public Article() {
Url = new ArrayList<String>(); date = new ArrayList<Integer>();
Site = new ArrayList<String>(); Headline = new ArrayList<String>();
Content = new ArrayList<String>(); Url = new ArrayList<String>();
} Site = new ArrayList<String>();
Content = new ArrayList<String>();
private ArrayList<Integer> date; }
private ArrayList<String> Headline;
private ArrayList<String> Url; private ArrayList<Integer> date;
private ArrayList<String> Site; private ArrayList<String> Headline;
private ArrayList<String> Content; private ArrayList<String> Url;
private ArrayList<String> Site;
public void setDate(int num) { private ArrayList<String> Content;
date.add(num);
} public void setDate(int num) {
date.add(num);
public void setHeadline(String head) { }
Headline.add(head);
} public void setHeadline(String head) {
Headline.add(head);
public void setUrl(String url) { }
Url.add(url);
} public void setUrl(String url) {
Url.add(url);
public void setSite(String site) { }
Site.add(site);
} public void setSite(String site) {
Site.add(site);
public void setContent(String content) { }
Content.add(content);
} public void setContent(String content) {
Content.add(content);
}
public int getHowManyData() {
return Headline.size();
} public int getHowManyData() {
return Headline.size();
public int getDate(int num) { }
return date.get(num);
} public int getDate(int num) {
return date.get(num);
public String getHeadline(int num) { }
return Headline.get(num);
} public String getHeadline(int num) {
return Headline.get(num);
public String getUrl(int num) { }
return Url.get(num);
} public String getUrl(int num) {
return Url.get(num);
public String getSite(int num) { }
return Site.get(num);
} public String getSite(int num) {
return Site.get(num);
public String getContent(int num) { }
return Content.get(num);
} public String getContent(int num) {
return Content.get(num);
}
} }

2
CAu_NLP_2019/src/main/java/Result.java → CAu_NLP_2019/src/main/java/com/nlp/Result.java
View File

@ -1,3 +1,5 @@
package com.nlp;
public class Result { public class Result {
private String tfidf; private String tfidf;

94
CAu_NLP_2019/src/main/java/TFIDF.java → CAu_NLP_2019/src/main/java/com/nlp/TFIDF.java
View File

@ -1,46 +1,48 @@
import java.util.List; package com.nlp;
public class TFIDF { import java.util.List;
/**
* @param doc list of strings public class TFIDF {
* @param term String represents a term /**
* @return term frequency of term in document * @param doc list of strings
*/ * @param term String represents a term
public double tf(List<String> doc, String term) { * @return term frequency of term in document
double result = 0; */
for (String word : doc) { public double tf(List<String> doc, String term) {
if (term.equalsIgnoreCase(word)) double result = 0;
result++; for (String word : doc) {
} if (term.equalsIgnoreCase(word))
return result / doc.size(); result++;
} }
return result / doc.size();
/** }
* @param docs list of list of strings represents the dataset
* @param term String represents a term /**
* @return the inverse term frequency of term in documents * @param docs list of list of strings represents the dataset
*/ * @param term String represents a term
public double idf(List<List<String>> docs, String term) { * @return the inverse term frequency of term in documents
double n = 0; */
for (List<String> doc : docs) { public double idf(List<List<String>> docs, String term) {
for (String word : doc) { double n = 0;
if (term.equalsIgnoreCase(word)) { for (List<String> doc : docs) {
n++; for (String word : doc) {
break; if (term.equalsIgnoreCase(word)) {
} n++;
} break;
} }
return Math.log(docs.size() / n); }
} }
return Math.log(docs.size() / n);
/** }
* @param doc a text document
* @param docs all documents /**
* @param term term * @param doc a text document
* @return the TF-IDF of term * @param docs all documents
*/ * @param term term
public double tfIdf(List<String> doc, List<List<String>> docs, String term) { * @return the TF-IDF of term
return tf(doc, term) * idf(docs, term); */
public double tfIdf(List<String> doc, List<List<String>> docs, String term) {
} return tf(doc, term) * idf(docs, term);
}
}
}