Unlike traditional search, InfoBot inherits GPT's understanding of every language on the Internet. So even if you index your documentation in Dutch, it can respond in the language of the question. Asking questions in a language other than what you've trained it to may not be as effective at this point as it's less likely to find the best context from our embedding index. We are working to improve this! You can train 2 InfoBots in 2 languages so that you always get your questions answered in both languages.

It's a bit technical, but here's a quick overview. We use OpenAI's embedding and ChatGPT APIs and vector databases to store our index. All included documentation is cleaned up and divided into smaller chunks and labeled by source. We then use the GPT embedding API to generate a vector representation of each chunk and store it in our vector db index. When a user asks a question, we convert it into an embedding and perform an advanced semantic search to look for matches to the user's question. We then take the most relevant chunks, include them as context along with the original question and use the ChatGPT API to generate an answer in markdown format which we then convert to HTML and display to the user.

Yes, you can enable scheduled resource updates for dynamic resources such as URLs, RSS feeds, and Sitemaps. Depending on your plan, you can do this monthly, weekly or daily.

When we train your bot using different sources, we calculate how many pages were imported. This corresponds to a web page url, or a page of a PDF or other document file. For example, if we import your website through a sitemap, it could contain hundreds of pages. For long document files, we calculate a page as roughly 4000 characters of cleaned raw text. Different plans have different page limits depending on your needs.

Some companies have different industries where the InfoBots need to be trained on different documents. By deploying multiple InfoBots, our InfoBots can become more specialized and provide better answers. Another use case could be that you have 1 InfoBot for your customers and 1 for your staff, of course your customers may not know the same as your staff, so you need 2 InfoBots trained on different information.