Access Bot Training:
Navigate to the Bot Training section in Pest AI.
Enter the URL:
Provide the full URL, including the https:// prefix.
Choose a Crawling Mode:
Select one of the three URL crawling modes (details below).
Fetch and Crawl URLs:
Allow the URLs to be fetched and processed.
Train Your Bot:
Select the desired URLs from the list and hit Train Bot.
Trained URLs will be added to the Uploaded Links Table with their status.
Review and Wait:
Wait for all URLs to complete training before using the bot.
Exact URL:
Crawl and train the bot on the specific URL provided.
Best for precise training with targeted content.
Steps:
Choose Exact URL.
Enter the URL and click Get Data.
Once processed, the bot is trained, and the URL is added to the Uploaded Links Table.
All URLs in This Domain:
Train the bot with all pages from a specified domain.
Ideal for covering broader site content.
Steps:
Choose All URLs in This Domain.
Enter the domain and click Get Data.
Review the list of URLs fetched, select the relevant ones, and click Train Bot.
New and existing pages will appear in separate lists:
New Pages: Not yet trained.
Existing Pages: Previously trained, can be refreshed.
All URLs with This Path:
Train the bot with pages containing a specific path in their URL.
Follow the same steps as for All URLs in This Domain.
Trained URLs are displayed in the Uploaded Links Table, showing their training status:
Getting Data: The bot is actively training or refreshing data.
Trained: Training is successful, with the timestamp of the last refresh.
Failed: Training failed; you can retry or delete the URL.
Additional Tips:
URLs take time to appear in the table; ensure all URLs are processed before use.
Regularly review and update the Uploaded Links Table to maintain relevant and accurate data.
Remove outdated links to improve the bot’s performance and response quality.
