Tool: Query Document Collection
The following showcase presents the Searchable Document Collection - Query and Searchable Document Collection - Add document or Searchable Document Collection - Add Website document tool. The tools allows you to add documents to a collection and a user can ask questions about this documents.
Please note that this tool has been created by the capabilities of AnySolve. You can create a tool that does a similar tool easily with the tool designer. Similar Tools: Create a Python script, Write and further improve a newspaper article, Write a paper, Create a HTML document
Introduction Video
Adding the document collection
First we need to build our knowledge base. For this we need to upload documents. In this example we add information pages about AnySolve so that we can later ask questions about AnySolve. The FAQ page of AnySolve https://docs.anysolve.ai/faq/
is added as faq
in the collection name anysolve
. The collection name
is important as all documents that should be queried together need to have the same collection name
.
Afterwards we add some more documents from the AnySolve website that have information about how to use AnySolve.
Please note: The document lifetime is specified in the tool description.
New: Add website document to the document collection
The tool Searchable Document Collection - Add Website document allows you to add a website by providing the URL
to a collection.
Querying the document collection
Now that we have a document collection we can query the knowledge base we have built. We can ask how can i create a html page with anysolve
. The document collection parameter must be set to the collection which we named anysolve
. We can also set the quality paramter which decides how well the quality of the tool is. The lower the quality the cheaper the tool run and the faster the result.
After a click on Run tool
we get the result.
The result shows the result as JSON format with answer
, source
, document_name
and text_excerpt
.
Conclusion
The tool allows you to add a document collection and queries the collection while getting a meaningful response.