search

leifeld / dna

Discourse Network Analyzer (DNA)
126 stars 41 forks source link
content-analysis discourse network network-analysis policy-debates text text-annotation

readme

Discourse Network Analyzer (DNA)

The Java software Discourse Network Analyzer (DNA) is a qualitative content analysis tool with network export facilities. You import text files and annotate statements that persons or organizations make, and the program will return network matrices of actors connected by shared concepts.

DNA 3.0 was first released on 12 June 2022. It constitutes a major rewrite from the previous version DNA 2.0 beta 25. DNA 3 comes with many new features and improvements. The release page contains all the details (scroll to version 3.0.7 for the first DNA 3 release).

build-check-test

Installation of DNA

Starting DNA

To use DNA, simply download the latest .jar file under "Releases" on the right along with the sample.dna file, a toy database you can load in DNA (password: sample) to play with the software. You can double-click on the .jar file to open DNA. No installation is required.

If your system does not want to start DNA with a double-click on the .jar file, you can also open it from the terminal. To do so, navigate to the directory where the .jar file is stored on your computer using the cd command, for example cd ~/Documents/. Once there, type in something like this, with the .jar file corresponding to the file version you downloaded:

java -jar dna-3.0.11.jar

Along with DNA, it is recommended to download visone, which can be opened in the same way as DNA. You can open .graphml files from DNA's network export in visone.

Java installation and version check

DNA was written in Java and requires Java 11 or higher on your system. You can check if you have Java >= 11 on your system by opening the terminal of your operating system (e.g., typing cmd in your Windows start menu or using the terminal app on MacOS) and typing:

java --version

If this indicates a version below 11 (or 1.11), installing the latest version of Adoptium Temurin OpenJDK or any other Java >= 11 is recommended before you proceed. Once installed, restart your system and check the version again.

MacOS

MacOS users may need to give the program authorization to be executed by opening the system settings and activating "Anywhere" rather than "App store and identified developers" under "Privacy & Security" -> "Security" -> "Allow applications downloaded from" (see here for instructions).

Some MacOS users reported issues with opening files. These issues should have been fixed in version 3.0.11.

You can also browse the issue tracker (including closed issues) and the commit messages for more information on MacOS issues. Get in touch via the issue tracker or Matrix (infos below) if you are unable to solve these issues.

Compiling from source using Gradle

If you require the latest (non-release) version of the DNA jar file from GitHub, you can clone the git repository to your computer and execute ./gradlew build on your terminal or command line. This will build the jar file, the rDNA R package, and the bibliography, and store them in the build/ directory of the cloned repository. If you only want to build the jar file, you can also execute ./gradlew :dna:build (omit ./ on Windows).

Alternatively, if you need the latest non-release version, you can try to download the latest artifact from the build process under GitHub Actions by clicking on the latest build and scrolling down to "Artifacts". However, it is usually recommended to use the most recent release version.

rDNA 3.0: Connecting DNA to R

The R package rDNA connects DNA to R for data exchange and analysis.

rDNA offers functionality such as:

The old 2.0 version additionally supported data exchange between DNA databases and rDNA by adding or retrieving documents, statements, attributes, statement types, coders etc. The 3.0 version does not support most of this data exchange functionality yet, so if you need to batch-import documents, for example, use DNA 2.0 first and then import everything into a DNA 3.0 database for further analysis (note that you cannot import from 3.0 into 2.0). More functions for data exchange and analysis will be added in the future.

To install the new rDNA 3 directly from GitHub, try the following code in R:

# install.packages("remotes")
remotes::install_github("leifeld/dna/rDNA/rDNA@*release",
                        INSTALL_opts = "--no-multiarch")

Note that the package relies on rJava, which needs to be installed first.

Installation of the old rDNA 2.1.18

For data management, you may still want to use the old rDNA 2.1.18 with DNA 2.0 beta 25. You can install the package directly from GitHub as well. However, you will need to download the correct JAR file and store it either in your working directory or (recommended) in the library path of the installed R package in the "extdata" subdirectory. The following code can do this for you:

# install.packages("remotes")
remotes::install_github("leifeld/dna/rDNA@v2.0-beta.25",
                        INSTALL_opts = "--no-multiarch")

# find out where to store the JAR file
dest <- paste0(dirname(system.file(".", package = "rDNA")),
               "/extdata/dna-2.0-beta25.jar")

# download JAR file and store in library path
u <- "https://github.com/leifeld/dna/releases/download/v2.0-beta.25/dna-2.0-beta25.jar"
download.file(url = u, destfile = dest, mode = "wb")

Documentation and community

Support the project

Please consider contributing to the project by:

Some suggestions of new functionality you could add via pull requests: