Guides
This section provides guidance on features, solution walkthrus, and other topics.
Getting Started
Kick off your journey with Tisane by learning how to quickly set up and start using Tisane:
Supported Features
- List of Supported Languages
- Full List of Supported NLP Functions
- Feature: Entity Extraction
- Feature: Topic Extraction
- Feature: Sentiment Analysis
- Feature: Built-in Translation
- Does Tisane Support My Programming Language?
- What does Tisane do when problematic content is detected?
- Algospeak and Adversarial Text Manipulations
- Explainability and Transparency
Types of Problematic Content
This section covers the concepts relevant to detection of concept of interest to moderators or, in law enforcement applications, of investigators.
- Personal Attacks/Insults/Cyberbullying
- Bigotry and Hate Speech
- Attempts to Establish External Contact
- Sexual Advances
- Adult-Only Content
- Allegations
- Criminal Activity
- Mental Issues and Suicidal Ideation
- Profanities
- Data Leaks
- Contentious Content
- Disturbing Content
- Meaningless Content
- Assertion of Social Hierarchy
- Spam
- Tags
- 2-Factor Moderation
How-to Topics
Guides to help you implement and fine-tune Tisane for various content moderation scenarios:
- Filtering By Keyword Blocklists
- Locating a Text Snippet Using Location Data
- Moderating Live Chat
- Moderating Usernames
- Detecting Threats
- Detecting Problematic Content
- Detecting Names of Countries
- Processing Competitive Language In Game And Sports Chats
- Considering Previously Processed Messages
Distribution and Deployment
- What Happens When My Monthly Quota Is Exceeded?
- On-Premise Deployment
- Can Tisane Be Embedded In My Application Deployed On A Standalone Machine?