Website
Last updated
Last updated
The Website data source in Epsilla allows users to import and manage content directly from webpages, making it ideal for dynamic or continuously updated information. This type of data source is optimal for retrieving data from websites, blogs, and other web-based platforms into the knowledge base.
To begin, choose Website as the data source to load content from webpages.
Click Continue to proceed.
Provide a Knowledge Base Name. The name should begin with a letter or an underscore (_), and can contain only letters, digits, underscores, and whitespaces. Enter your desired name in the input box, such as MyWebsiteKnowledge.
In the Webpage URLs section, input the publicly accessible URLs you want to extract data from.
The URLs must start with http:// or https://.
You can add a Single Webpage manually, Crawl a webpage for subpages, or add multiple webpages at once.
For example, you can manually enter:
To crawl a webpage, click the Crawl webpage button. In the dialogue that appears:
Input the base URL, such as https://epsilla-inc.gitbook.io/.
Set the Max number of pages to crawl (e.g., 100).
Click Search to locate subpages. A list of subpages will be displayed.
Select the pages you want to include in the knowledge base and click Add.
You can also use the Add multiple webpages option to input multiple URLs at once, separated by new lines.
Once you've added the desired URLs, click Create to begin processing the data.
Epsilla will automatically retrieve the content from the pages, chunk it into manageable pieces, and embed it into vectors. You can monitor the progress during this step.
You can inspect the processed data (chunks) at the tab.
