What is the core function of the Tisane API and what problem does it address?
The Tisane API provides natural language processing (NLP) for user-generated content, with a primary focus on automatic content moderation. It detects issues like personal attacks, hate speech, profanities, and PII exfiltration, solving the problem of identifying and explaining harmful or problematic text at scale. Its explainability feature provides reasons for detections, making it compliant with standards like the DSA.
In what situations and for what specific uses is the Tisane API employed?
Tisane is used for detecting and analyzing a wide range of content issues in user-generated text, including cyberbullying, hate speech, criminal activity, sexual harassment, and adult-only references. It is applied to parse usernames, detect adult advertising in profiles, extract topics and entities, and translate obfuscated or dirty language across 35 languages.
Who is the intended audience or user base for the Tisane API?
The target audience is developers and trust & safety teams who need to integrate content moderation and text analysis into their applications. This includes global teams requiring translation of offensive content and organizations needing scalable moderation solutions.
What key features, deployment options, and integrations does the Tisane API offer?
Key features include explainability, the ability to counter algospeak and adversarial text, flexibility for context-specific processing, and built-in automatic translation. It is available as SaaS, On-Premise, and Embedded deployments. Integrations include plugins for Slack, Webex, Maltego, Minecraft, and Zapier.
What is Tisane?
Tisane is a Developer & AI Platform that offers a natural language processing API focused on automatic content moderation for user-generated content. It detects issues like hate speech, personal attacks, and profanity, and provides explainable results in 35 languages.
5 of 6 research questions are answered for this product. The rest need source evidence we have not collected yet, so they are left unanswered rather than guessed.