A New Research Module for Identifying Both Toxic and Prosocial Interactions in Online Discourse
Toronto, Ontario – [06/17/2025] – We are pleased to announce the launch of the new Civility Analyzer module in Communalytic, a no-code computational social science platform developed by the Social Media Lab at Toronto Metropolitan University. This module is designed to assess the tone and quality of online interactions by automatically identifying both toxic and prosocial content in text-based datasets. It integrates two leading machine learning models: Google’s Perspective and the open-source Detoxify model.
The Civility Analyzer extends beyond traditional toxicity detection by also identifying constructive, prosocial exchanges, offering researchers an accessible, data-driven approach to studying both prosocial and harmful behaviours across various social media platforms. The module has been available in beta for the past six months and is now fully tested and officially released on both the EDU and PRO versions of Communalytic.
Advancing Social Media Research and the Study of Online Civility
The Civility Analyzer helps researchers, journalists, community managers, and platform designers identify posts containing threats, insults, harassment, or profanity. It also detects prosocial communication, including support, cooperation, and kindness. This dual capability offers a balanced view of digital conversations, capturing both harmful and constructive interactions. The analyzer calculates toxicity scores (such as Toxicity, Insult, and Threat) and prosocial scores (like Compassion, Curiosity, and Respect) for each post. These scores, ranging from 0 to 1, estimate the likelihood that readers would perceive an interaction as toxic or prosocial. For example, a toxicity score of 0.7 suggests that 7 out of 10 people might see the post as toxic.
By releasing this module, our team aims to make it easier to analyze social media conversations, uncover discourse trends, and understand the drivers behind online interactions, all while supporting healthier digital spaces. The Civility Analyzer provides both detailed post-level insights and broader trend analysis, helping to build more respectful and inclusive platforms. Whether used for content moderation, social research, or developing strategies to encourage prosocial behaviour, this tool provides a practical, scalable way to assess the tone of online communities.
Integrated, Multilingual Analysis
The module supports multiple languages, including Arabic, Chinese, Czech, Dutch, English, French, German, Hindi, Hinglish, Indonesian, Italian, Japanese, Korean, Polish, Portuguese, Russian, Spanish and Swedish (via Google’s Perspective) and English, French, German, Italian, Spanish, and Portuguese (via Detoxify).
Users can either collect data from popular social media platforms or upload datasets directly into Communalytic, where each post is automatically evaluated across a range of toxicity and prosocial dimensions. The system provides visual summaries, including time-series charts and distribution graphs, and allows for the export of detailed scores for further analysis and review.
For more details, see our Learn More, our FAQs, and our Tutorials
About Communalytic
Communalytic is a no-code computational social science research tool for studying online communities and public discourse on social media. It is designed to provide researchers, journalists, and students with essential resources and infrastructure for conducting independent, public-interest research. It has a full suite of easy-to-use social media data collectors and analyzers – no coding is required. Users can bring their own data or use one of Communalytic’s various social media data collectors to collect data from platforms such as Bluesky, Mastodon, Reddit, Telegram, X (formerly Twitter), and YouTube.
There are two versions of Communalytic. Each is designed for different purposes and different sets of users:
- Communalytic EDU is designed to help students learn about social media data analytics.
- Communalytic PRO is designed for academic researchers and journalists and is ideal for large-scale research projects.
About Communalytic’s Data Analyzers Modules
Communalytic also comes with a set of built-in data analytics modules, including a:
- Civility Analyzer that can identify antisocial and prosocial interactions in a dataset using the latest machine-learning models (Perspective API and Detoxify),
- Sentiment Analyzer that can calculate sentiment polarity scores to determine whether the text in a dataset expresses a positive, negative or neutral sentiment,
- Topic Analyzer that can automatically group social media posts that are semantically similar to identify latent topics in a dataset (i.e., abstract topics that may not be directly observable from just reading the posts),
- Network Analyzer that can generate and visualize various types of networks in a dataset, including signed and unsigned communication networks, as well as link-sharing networks. A signed network is one where the nodes and edges carry additional information such as weights (i.e., toxicity and prosocial scores or sentiment scores)
These data analyzer modules can automatically:
- Detect antisocial (Toxicity, Insults, Threats … ) and prosocial interactions (Compassion, Curiosity and Respect …) in any text-based dataset,
- Assess sentiments in online discourse (i.e., opinion mining),
- Group together social media posts that are semantically similar and identify latent topics, uncovering hidden communities of users who share an interest in a topic but may not know each other or have ever communicated with one another.
- Find out who talks to whom, who shares whose contents, who shares the same links or resources, etc…
When used together, these analytical modules can be used to study online communities and influencers, map shared interests among community members, study the spread of misinformation and disinformation, and detect signs of possible coordination among seemingly disparate actors.