Gephi 매뉴얼 - Gephi maenyueol

Welcome to Gephi! Gephi is an open-source software for visualizing and analysing large networks graphs. Gephi uses a 3D render engine to display graphs in real-time and speed up the exploration. You can use it to explore, analyse, spatialise, filter, cluterize, manipulate and export all types of graphs.

Getting Started

New to Gephi? These guides are for you. The following pages should introduce you to the Gephi project, basic features and installation.

• Goals and Principles of the project
• System Requirements, Installation, Release Notes
• User interface description
• Exported File Formats

Official Tutorials

Gephi is really easy to handle if you learn the basics. Let's follow these tutorials to quickly manage the main features!

• How to Import Spreadsheet (Excel) Data // video
• How to Import Dynamic Data

Popular Tutorials by the Community:

• Gephi tutorials maintained by Clément Levallois
• Gephi – Introduction to Network Analysis and Visualization
• Using Netvizz & Gephi to Analyze a Facebook Network
• Getting Started With The Gephi – My Facebook Network
• Dynamic Networks in Gephi: From Twapperkeeper to GEXF
• Generating graphs of retweets and @-messages on Twitter using R and Gephi
• Text Network Analysis
• Visualize keywords and landing pages from Google Analytics

Various Tutorials in Video:

• How to group elements, apply filters, use the timeline, and receive a live graph stream.
• Facebook Network Analysis
• Text Network Analysis
• Proliferating Information through Communities
• PubMed-derived Citation Data

Non-English Tutorials:

In French:

• Analyser graphiquement son réseau facebook
• Cartographier vos informations avec Gephi
• Introduction aux graphes avec Neo4j et Gephi
• Exploration visuelle d’un netlinking, réseaux de données, et plus encore… (Part I)

In Spanish:

• Taller Gephi
• CAMON MADRID: Taller de Visualización de Datos

In Chinese:

• 介绍 Gephi 的各项主要操作方法。

Datasets

Let's download and try some datasets available on the wiki, like C.elegans brain network or the web mapping study EuroSiS for the European Union.

Practical cases

A Twitter tag, #madewithgephi, has been adopted to tell when Gephi was used. You can find a compilation of (some! of) these works on the forum.

Facebook group

Many experienced users can provide help on this group.

Forum

New users are welcome to post questions at the Gephi Support Forums. If you have a technical problem and have read the frequently asked support questions, and still can't find the answer, register then click the "[Create New Topic]" link and fill out the form to pose your question to forum members.

Wiki

The Community portal centralizes all information for a better life :)

Papers

Published papers highlight Gephi usages and techniques. We have a collection of papers in which reseachers use Gephi, or explaining the underlying ideas.

Roadmap

Discover our roadmap and submit your whishes in this wiki page.

History

Find out who is behind the project, the Gephi Team, and our story.

FAQ

A list of common problems and their solutions can be found in the Support Forums. Technical issues only may be adressed in our Tech

. You can ask your technical questions here.

System Requirements Specification Document

A document contributed by Konstantinos Varvoutas.

데이터 시각화 방법 정리

1. 빅카인즈에서 다운 받은 Raw 데이터를 어느 정도 필터링 하여 단어 분할한 후에 Gephi로 Import 한다.

2018년 E스포츠 기사들이다.

2. Statistics를 수행한다.

• Average Degree
• Modularity
• Eigenvector Centrality

3. Filters 를 설정한다.

• Degree Range
• Partition ( 이후 과정에서 설정 )

4. Filter 를 클릭하면 아래와 같이 그래프가 출력된다.

• 위 그래프는 각 노드의 Degree(엣지 개수)가 400개 이상인 노드들만 모아진 것이다.
• Arial Bold, 32 를 클릭하여 폰트를 한글 폰트로 설정한다. ( 한글 폰트로 바꾸지 않으면 한글 텍스트가 깨진다. )

5. Appearance 를 설정한다.

• EigenVector Centrality의 크기 에 따라 노드의 크기 를 크게 해준다.

• Apply 를 클릭했을 때의 결과

• Modularity 에 따라 색상을 다르게 해준다.

• Apply 를 클릭했을 때의 결과

6. Layout 을 설정한다.

• Force Atlas 2
• Scaling : 1000

• Run 클릭 했을 때의 결과
• 여기에서 빨간 박스 를 클릭하면 각 노드의 이름이 나타난다.
• 어느 정도 형태가 변하지 않는다면 Stop을 클릭한다.

7. 결과

• 이제부터는 Modularity로 나뉜 영역의 의미와 각 노드들의 의미를 분석한다.

★. 겹쳐져있는 노드 떼어내기

• 이렇게 겹쳐져 있는 노드들은 하나하나 드래그 하여 떼어내는 것도 방법이지만

• Label Adjust 기능을 통해 각 노드들을 떼어내준다.

★. 각 Modularity 만 관찰하기

• Filter 구성
  • Main Filter : Partition ( Modularity Class )
  • Sub Filter : Degree Range
• 각 Partition을 클릭하면 해당 Partition 만의 그래프를 출력한다.

이제 분석을 시행하면서 데이터 수정을 반복한다.