Our Portfolio

Nullam id dolor id nibh ultricies vehicula ut id elit.

4chan & 8chan Word Embeddings

Online anti-Semitism across platforms (TGTR-7)

Abstract Authors Abstract We created a fine-grained AI system for the detection of anti-Semitism. This Explainable AI will identify English and German anti-Semitic expressions of dehumanization, verbal agression and conspiracies in online social media messages across platforms, to support high-level … Read More

The sexist narrative on alternative social media dissected (TGTR-6)

Abstract Authors Abstract Toxic language use on fringe social media platforms and image boards (4chan, 8kun, GAB, etc.) is being increasingly well-documented, with many studies describing the racist, nationalist and antisemitic rhetoric that are flooding these boards in particular. Yet, … Read More

Geenstijl.nl embeddings

Onze Echokamers: likes onder de loep (TGTR-5)

Abstract Authors Abstract Textgain onderzocht hoe echokamers automatisch in kaart kunnen worden gebracht aan de hand van publieke data van de twitter-accounts van nieuwssites, influencers en politici. In dit artikel beschrijven we de huidige stand van zaken in het Nederlandse … Read More

GeenStijl.nl embeddings (TGTR-4)

Abstract Authors Abstract We collected over 8M messages from the controversial Dutch websites GeenStijl and Dumpert to train a word embedding model that captures the toxic language representations contained in the dataset. The trained word embeddings (±150MB) are released for … Read More

QAnon: Spreading Conspiracy Theories on Twitter

Abstract Authors Abstract From 1st October to 5th November 2020, Textgain analyzed half a million Twitter messages related to QAnon conspiracy theories, using  our Natural Language Processing (NLP) technology. This report outlines the results of the findings of our quantitative … Read More

Using a Personality-Profiling Algorithm to Investigate Political Microtargeting

Abstract Authors Abstract Political advertisers have access to increasingly sophisticated microtargeting techniques. One such technique is tailoring ads to the personality traits of citizens. Questions have been raised about the effectiveness of this political microtargeting (PMT) technique. In two experiments, … Read More

Profanity & Offensive Words (P​OW​): Multilingual fine-grained lexicons for hate speech (TGTR-3)

Abstract Authors Abstract The POW lexicons are a steadily growing, interpretable NLP resource for online hate speech detection. They are currently available in English, German, French, Dutch and Hungarian, capturing thousands of verbal expressions of abusive, aggressive, dehumanizing, discriminatory, offensive … Read More

MAL NLP Lexicon: Melancholy, Anxiety & Loneliness during lockdown (TGTR-2)

Abstract Authors Abstract We have created a new practice-based NLP resource for monitoring mental health on social media, in particular brooding. The resource is currently available for Dutch and captures 2,000+ expressions of anger, fear and sadness, along with various … Read More

4chan & 8chan embeddings (TGTR-1)

Abstract Authors Abstract We have collected over 30 million messages from the publicly available /pol/ message boards on 4chan and 8chan, and compiled them into a model of toxic language use. The trained word embeddings (±0.4GB) are released for free … Read More

Multilingual Cross-domain Perspectives on Online Hate Speech

Online hatred of women in the Incels.me forum – Linguistic analysis and automatic detection

Automatic detection of cyberbullying in social media text

Summary Authors Summary While social media offer great communication opportunities, they also increase the vulnerability of young people to threatening situations online. Recent studies report that cyberbullying constitutes a growing problem among youngsters. Successful prevention depends on the adequate detection … Read More


Arabic Dialect Identification

Dutch Word Embeddings


GDPR Anonymization Tool


Text-Based Age and Gender Prediction for Online Safety Monitoring

Automatic Detection of Online Jihadist Hate Speech

Summary Author Summary We have developed a system that automatically detects online jihadist hate speech with over 80% accuracy, by using techniques from Natural Language Processing and Machine Learning. The system is trained on a corpus of 45,000 subversive Twitter … Read More