Tutorials

For Teaching & Learning

EDU Login
EDU Sign Up - It's Free!

For Research

PRO Login
PRO Sign Up - Free 7-Day Trial

New to Communalytic? We have prepared a number of tutorials to help you get started.

If you are using Communalytic in an academic publication, please cite us as: 

  • Gruzd, A., & Mai, P. (2022). Communalytic: A Research Tool For Studying Online Communities and Online Discourse. Available at https://Communalytic.com

There are two versions of Communalytic: EDU and PRO. Each version is hosted on its own dedicated server with its own account creation and sign-in processes. Users of Communalytic can share datasets with other users who are using the same version of Communalytic (i.e., EDU users with EDU users and PRO with PRO). (For more info, see Section 9: Data Management)

The Toxicity Analysis module in Communalytic can conduct toxicity analysis (via Google Perspective API) on text in the following languages – Arabic, Chinese, Czech, Dutch, English, French, German, Hindi, Hinglish, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish.

  • Tutorial: Obtaining a Perspective API Key
    • Troubleshooting tips for obtaining a Perspective API Key
      • The Google account used to obtain a Google Perspective API Key can be different from the Google account you used to create your Communalytic account. 
      • In some instances, Google might not allow you to create a Google Cloud project with your academic/institutional email. If that is the case, you will need to use a Google account ending with @gmail.com.
  • Tutorial: Toxicity Analysis in Communalytic

The Sentiment Analysis module in Communalytic can conduct sentiment analysis on text in the following languages: English, French, German and Russian using one or more of the following three popular sentiment analysis libraries: VADER (EN), TextBlob (EN, FR, DE) and Dostoevsky (RU).

  • Posts in French or German will only be analyzed by TextBlob.
  • Post in Russian will only be analyzed by Dostoevsky.
  • Posts in English will be analyzed by both VADER and TextBlob. Researchers with a predominantly English language dataset will have the option to inspect conflicting polarity scores generated by these two different sentiment analysis libraries (VADER and TextBlob) and decide which library is better suited/more accurate for analyzing their particular dataset.

The Network Analyzer module in Communalytic can automatically generate and visualize various types of networks including communication networks, two-mode semantic networks, link-sharing networks and word co-occurrence networks.

Types of networks that can be automatically generated by Communalytic

  • Reply-To Network: Account-to-Account (Reddit, Twitter, Telegram groups) 
    • This communication network shows who replied to whom. 
  • Retweet Network: Account-to-Account (Twitter only)
    • This communication network shows who retweeted whom.
  • Two-Mode Semantic Network*: Account-to-Named Entity (Reddit, Twitter, CrowdTangle, Telegram channels & groups)
    • This semantic network shows which account mentioned what ‘named entity’**. 
    • The named entity detection is based on an advanced Natural Language processing library called spaCy.  
  • Two-Mode Semantic Network: Account-to-Named Entity (Twitter only) 
    • This semantic network shows which account mentioned what ‘named entity’ and is based on Twitter’s automated annotation. This approach is faster than using spaCy, but may miss some named entities. 
  • Two-Mode Link Sharing Network: Account-to-Website (Reddit, Twitter, CrowdTangle, Telegram channels & groups) 
    • This ‘link sharing’ network shows which accounts in your dataset shared a link to the same website(s). 
  • Word co-occurrence network: Named Entity-to-Named Entity (Reddit, Twitter, CrowdTangle, Telegram channels & groups)
    • This network connects two or more ‘named entities’ mentioned in the same post(s).
    • The named entity detection is based on an advanced Natural Language processing library called spaCy. 

Creating a signed network in Communalytic

The Network Analyzer module in Communalytic is unique among network research tools in that it can generate and visualize so-called “signed” networks. A signed network*** is a network with edges that contains additional information such as positive or negative signs or scores (weights). To turn a network into a signed network in Communalytic, users have the option to run a couple of additional analyses (toxicity and/or sentiment) prior to creating a network representation of their dataset. The resulting toxicity scores and/or sentiment polarity scores would be added as weights to edges in the network and visualized for easier exploration and analysis. This feature can be used to identify and visually highlight interactions of interest (e.g., anti-social interactions) within the network so that they may be examined in more detail.

In addition, if a user is working with Twitter data and completed a Bot detection analysis prior to creating a network representation of their dataset, the resulting bot probability scores would be added as weights to nodes in the network and visualized for easier exploration and analysis. This feature can be used to identify and visually highlight interactions of interest (e.g., Twitter accounts that might be bots) within the network so that they may be examined in more detail.

Additional resources 

Learn more about signed networks

Definitions

* A two-mode semantic network is a graph that connects two types of nodes, where one of the node types represents social actors (accounts) and the other node types represents semantic concepts (operationalized in Communalytic as named entities). A connection from a social actor to a semantic concept in such a network usually implies some form of endorsement, association or affiliation between the two nodes. The exact interpretation of social actors, semantic concepts and network connections will depend on the available data (including any metadata) and research questions that the researcher would like to answer.

** Named entities can be people, organizations, locations, products, etc. as detected by an advanced Natural Language Processing library called spaCy. 

*** A signed network is a network with edges that contains additional information such as positive or negative signs or scores (weights).