Learn Tools and Techniques

This tool helps you design advanced interactive graphs.

Find more information about the programme here.

This package detects coordinated link sharing amongst accounts on Facebook to potentially unveil coordinated inauthentic behaviour (CIB). Output can be written also directly to a GRAPHML format for use in visualization (e.g. Gephi).

See tutorials page here.

This visualisation package helps you map out social networks to understand the most active and highly connected users. It can be used to identify coordinated campaigns.

See learn how page here.

This is a great YouTube video about using Gephi for network analysis.

This Python package will automatically run a sentiment analysis of English text. 

See the official quick start guide here.

Further documentation can be found here.

HateSonar allows you to detect hate speech and offensive language in text, without the need for training. 

See the documentation on GitHub here to get started.

Use CrowdTangle’s Meme search to search through Facebook page and group posts by text within images (e.g. posts that include images with text signs).

See our methodology on meme analysis to see how you can put this tool to work.

An online webpage that allows you to get a score for Twitter accounts on how ‘bot-like’ their activity is. It also has an API for advanced users and offers links to databases of known bots. 

For more information, visit the homepage.

This online tool allows you to enter search terms to live search Twitter or upload a csv file with your own Twitter data, to visualize the spread of information, including identification of possible bots. The back end and front end are open source and available on Github for advanced users. 

A tutorial will be offered when you first visit the website (clear your cookies to be offered the tutorial again).

You can also use the drop-down Help > Tutorial.

A software (currently in beta-testing) for real-time detection of new coordinated campaigns. Links to Hoaxy for visualization. 

Obtaining a Twitter developer account to access the API is straightforward and can be applied for here.

An online tool for deep examination of photographs to help detect manipulation. 

Tutorial can be found on YouTube.

Tool that provides information on Twitter account activity, such as time of posts and repetition, to allow identification of possible bots. 

None

An advanced Twitter scraping and OSINT tool written in Python that does not use Twitter’s API, allowing you to scrape a user’s followers, following, Tweets and more without restrictions.

User friendly, even for beginners.

Many tutorials online, including on YouTube. Twint’s community refers specifically to these websites:

https://pielco11.ovh/posts/twint-osint/

https://null-byte.wonderhowto.com/how-to/mine-twitter-for-targeted-information-with-twint-0193853/

Web application that analyses the activity of specific Twitter accounts, such as numbers of tweets, followers, hashtags and following. Advanced features allow for identification of potential bots.

Find more information about the free version here.

Mentionmapp displays how a specific Twitter account is connected to other user profiles and hashtags in a visual network.

Watch a live demo here.

This verification tool allows you to perform reverse image search, fragment videos from various platforms into key frames and read video and image meta data.

It is available as a free plugin for Chrome and Firefox.

See our methodology on reverse image search to see how you can put this tool to work.

Similar to InVid, this extension is another useful tool for reverse image searching.

It is available as a free plugin for Chrome and Firefox.

See our methodology on reverse image search to see how you can put this tool to work.

An easy-to-use Python library for accessing and interacting with the Twitter API.

You need a Twitter developer account.

Find installation and get started tutorials here.

SentiStrength estimates the strength of positive and negative sentiment in short English texts. Other languages are available too and can be easily added.

SentiStrength is free for academic research only.

Tutorials can be found here.

TweetDeck allows for multiple advanced Twitter searches in columns across a researcher’s
screen. The columns update automatically, which is an excellent feature for active monitoring.

Requires no programming skills.

For more
information, see this TweetDeck guide from Bellingcat.

This AI technology allows researchers
to automatically detect and find specific objects of interest in a large collection of images.

Learn more about the tool here.

Pre-written code that is adaptable can be found on GitHub.

Vision API offers pre-trained machine learning models through REST and RPC APIs that can detect faces, emotions,
landmarks, crops, image properties, labels,
landmarks and logos in images, as well as web properties,
including where else the image exists on the
web.

Requires programming skills.

Find tutorials and sample codes here.

Gensim is a free open-source Python library, designed to process raw, unstructured digital texts using unsupervised machine learning algorithms.

Tutorials can be found here – and a great beginners guide here.

See the documentation on GitHub to get started.

A website that allows you to search open Telegram channels and chats. Views statistics, subscribers growth, forwards and mentions of channel.

None

An easy-to-use Python library for accessing and interacting with the Twitter API.

You need a Twitter developer account.

Tutorials can be found here.

See the documentation on GitHub to get started.

Datawrapper lets you show your data as beautiful charts, maps or tables without using any code.

Easy to use, no coding or design skills are needed.

Tutorials and training material can be found here.

An open source data visualisation framework, RAWGraphs works with tabular data.

Based on scalable vector graphics format, visualisations can be easily edited with vector graphics applications for further refinements or directly embedded into web pages.

Easy to use, no coding skills are needed.

Tutorials can be found here.

See the documentation on GitHub to get started.

A collection of charts made with Python, including their associate reproducible code.

The same useful overview is available for R too.

Requires coding skills.

Graphs are mostly created with Matplotlib and Seaborn but other libraries like Plotly are sometimes used too.

The website consists of Jupyter Notebooks that can be found on GitHub.

An easy-to-use Python package that provides an expressive and flexible API for scraping Instagram data.

Designed for seamless integration with Selenium and Pandas.

Find installation and documentation here.

Workbench is an open source platform that enables all stages of data journalism: getting data (including scraping), cleaning, analysing, visualizing, and sharing it.

No coding skills needed.

Here you can find more information about the platform’s features.

You can find useful tutorials here, starting with “An Introduction to Data Journalism”.

Rainette is an R package which implements a variant of the Reinert textual clustering method. It features simple or double clustering algorithms, plot functions to explore clustering results and utility functions to split a text corpus into segments.

Programming skills are needed.

See the documentation on GitHub to get started.

Tableau is a visual analytics platform that helps people to explore, visualise and manage data. Output options include multiple chart formats as well as mapping capability.

No coding skills needed. SQL knowledge is an advantage.

You can find useful tutorials and free training videos here.

This tool is a JavaScript-based option for creating web and mobile dashboards. FusionCharts gives ready-to-use code for all of the chart and map variations, making it easier to embed in websites even for users with limited programming knowledge.

FusionCharts is aimed at creating dashboards rather than just straightforward data visualization.

Programming skills are needed. Available upon subscription.

Helpful guides and resources to set up a successful dashboard can be found here.