GSAK-KIT

This project provides a plugin for graph analysis and visualization through which you can easily embed, visualize, and analyze your data on any website or web page.

Building the Project:

Prerequisits:

• Apache Maven 3.x
  
• Java 8
  
• Git (Optional)
  

Steps

• Download or clone Gsak-Kit from github into any convenient Directory (Path should not contain special Characters like spaces etc.).
  
• Go to Project root directory (Directory containing src, pom.xml)
  
• To add dependencies run the following commands in project root directory using command line.
  mvn install:install-file -Dfile=src/lib/gephi-toolkit.jar -DgroupId=gephi-toolkit -DartifactId=gephi-toolkit -Dversion=1.0 -Dpackaging=jar

mvn install:install-file -Dfile=src/lib/gephi-cw.jar -DgroupId=gephi-cw -DartifactId=gephi-cw -Dversion=0.0.1 -Dpackaging=jar

mvn install:install-file -Dfile=src/lib/uk-ac-ox-oii-sigmaexporter.jar -DgroupId=uk-ac-ox-oii-sigmaexporter -DartifactId=uk-ac-ox-oii-sigmaexporter -Dversion=1.0 -Dpackaging=jar
  
  
• To build the Project, execut the following commands:
  mvn compile
mvn clean package
  
• If compilation completed without any error, a target directory will be created inside project root directory.
  
• The target directory should contain a jar file with possible name gsak-kit-with-dependencies.jar.
  
• To deploy gsak-kit server, execute the following command inside target directory.
  java -jar gephikit-0.0.1-SNAPSHOT-jar-with-dependencies.jar ../configuration.json
  
• Open localhost:8282/ in any browser; A welcome message will be display.
  

Routes:

Routes define different graph operations that can be applied to the visulized graph. Standard operations mainly include graph Layouts, Filtration or combination of both.

Welcome Route

• Route: / It will display welcome message
  
  

Basic Graph

• Route: /gephi
  

• Parameter(optional and Data Source Specific):

• Elasticsearch Params Options: basicSettings={"selectedDataSource":"elasticsearch","searchValue":"KEYWORDS'","host":"localhost","port":"9300","clusterName":"elasticsearch","index":"indexName","type":"indexTypeName","documentsLimit":"1000"}
  

• Text File Params Options: basicSettings={"selectedDataSource":"file","filePath":"path_to_the_tweets_file"}
  

  • e.g. File as Data source:: localhost:8282/gephi?basicSettings={"selectedDataSource":"file", "filePath":"/etc/tweets.txt"}
    

Layouts

• Route: /layout
  
• Layout name: "name"=YifanHuLayout
  
• e.g. YifanHuLayout: localhost:8282/layout?layoutSettings={"name":"YifanHuLayout","distance":100, "iterations":100}
  
• e.g. FruchtermanReingold: localhost:8282/layout?layoutSettings={"name":"FruchtermanReingold","size":10, "iterations":100, "gravity":0.1}
  
• Possible Layouts: YifanHuLayout, FruchtermanReingold

Filters

• Route: /filter
  
• Possible Filter Options: pageRankThreashhold, nodeCentrailityThreashhold, neighborRangeThreashhold
  
• e.g. YifanHuLayout: localhost:8282/filter?filterSettings={"pageRankThreashhold":10,"nodeCentrailityThreashhold":10, "neighborRangeThreashhold":10}
  
  

KIT-Route

• Route: /gsakkit
  
• Possible Filter Options: pageRankThreashhold, nodeCentrailityThreashhold, neighborRangeThreashhold
  
• Description: This route bascially combines all three major functionalities of GSAK-KIT; i.e. Source selection, Layouts, and graph Filteration.
  
• e.g. Example: localhost:8282/gsakkit?basicSettings={"selectedDataSource":"file", "filePath":"/etc/tweets.txt"}&filterSettings={"pageRankThreashhold":10,"nodeCentrailityThreashhold":10, "neighborRangeThreashhold":10}&layoutSettings={"name":"FruchtermanReingold","size":10, "iterations":100, "gravity":0.1}
  
  

GSAK-KIT is divided into two sections.

The server side

• input
  Text data or the built-in graph
  
• output
  Processed graph
  
• Description
  The server side will do all the data processing, graph generation, analysis, filteration, etc. When the server is provided with text data, it generates a graph based on specific keywords or relationships that exist within the data. When the server is provided with the built-in graph, it will analyze and filter the graph to create another processed graph. The server specifies formats for the input data. From this, the server generates the graph and provides a way to write your own implementation code for the graph generator.
  
• Third Party Libraries
  For graph processing we'll use the GephiToolkit
  For java web server we'll use the Undertow
  For logging we'll use log4j2

The front end side

• input
  Processed graph data
  
• output
  Sigma graph
  
• Description
  The front end will provide a way for visualizing the graph generated by the server. The front end will also provide a way to use the graph services in a webpage.
  
• Third Party Libraries
  For front end visualization we'll use the SigmaJs
  
  The front-end visualization is completely independent of the backend graph generation/processing, it will enable users to design their own front-end layouts and to use graph services where they need.

License

The code is licensed under the Apache License Version 2.0

Questions or Suggestions

Email [email protected]
Issues https://github.com/project-spinoza/gsak-kit/issues