Training Your Chatbot
One of the most powerful features of Saylient AI is the ability to train your chatbot with custom data. This allows your chatbot to provide accurate and relevant responses based on your specific content and domain knowledge.
Introduction to Data Sources
Your AI chatbot can learn from various sources, enhancing its accuracy and ability to offer precise and informed responses to user questions. Saylient AI supports multiple types of data sources:
- Text Input
- PDF Documents
- Website Links
- RSS Feeds (including Substack)
To access the Data Sources section:
- Navigate to the "Data Sources" option in the left sidebar menu.
- You'll see a list of your current data sources (if any) at the top of the page.
Adding a New Data Source
To add a new data source:
- Scroll down to the "Add Data Source" section on the Data Sources page.
- You'll see a list of vertical tabs, each representing a different data source type.
- Click on the tab for the type of data source you want to add.
- The corresponding form for that data type will appear, allowing you to input the necessary information.
Types of Data Sources
1. Text Input
Directly input text content for your chatbot to learn from. This is useful for adding specific information or custom responses.
[Detailed instructions for adding text input will be provided in a separate section]
2. PDF Documents
Upload PDF files containing relevant information for your chatbot to learn from.
3. Website Links
Adding a website allows Saylient AI to crawl and extract information from the specified web pages.
4. RSS Feeds (including Substack)
Connect RSS feeds to keep your chatbot updated with the latest content from blogs, news sites, or Substack newsletters.
Managing Data Sources
In the Data Sources page, you can:
- View all your added data sources at the top of the page
- Check the status of each data source (e.g., "Trained")
- See when each data source was last updated
- View or delete existing data sources
Need Assistance?
If you're experiencing any difficulties with training your data source or have questions, don't hesitate to reach out to the Saylient AI support team at hello@saylientai.com.