/slashtag-server

Primary LanguagePython

slashtag-server

Web Assets:

Accounts

Main Applications

Social Media + Marketing:

Marketing + Positioning

  • name: SlashTag
  • tagline: Data Annotation for your Team, built right into Slack
  • description: Empower the experts on your team to build better machine-learning models and annotate your datasets without leaving Slack. Teams can collaborate to make highly accurate datasets or leverage our active learning systems to focus on labeling the most important data.
  • url: https://slashtag.dev
  • logo: ./media-kit/logo.svg

Design:

- Primary #FFBD59
- Secondary #599BFF
- Tertiary #D9E8FF
- Dark Background #5963ff

Types of annotations:

Data:

  1. Text (raw, tweets, slacks, etc..)
  2. Image
  3. Video
  4. Audio
  5. File

Types:

  1. Binary / Multi-Class
  2. Can we do span selection for NLP?
  3. Bounding Boxes?
  4. Free Text / Survey Like

Case Studies:

Pull in examples of building data-driven applications with SlashTag. Would be cool to create a live project that is a real-world example of how SlashTag can be used. Users can annote it usign SlashTag and then use the data to build a model (perhaps need 3+ users to agree so that it doens't get too noisy).

  1. Train own Reddit Sentiment Data From Scratch

    1. Compare against existing models

  2. Fine-tuning Training BERT model for domain specific uses

  3. Labeling CRM data as it flows into Slack

  4. Find other popular tutorials and fast-follow them.

Tools:

Design: