Overview

This process involves extracting and answering questions from an uploaded questionnaire (e.g. an request for proposal questionnaire), ensuring that each response is thorough and well-supported by documents available in the Knowledge Center. The model utilizes advanced natural language processing to review and integrate relevant information from various documents, creating comprehensive and precise answers.

Questionnaire input format

Please ensure that the uploaded questionnaire is an Excel file with a single column for the questions. The column header should be "Question" Adhering to this format is crucial. Below is an example of the required input file.

The answers are then presented in the following formats:

1. as an output in a chat interface, featuring clickable links to the source materials,

2. as a comprehensive list in a final Excel document. This document compiles all the responses alongside their respective document references for easy review and verification.

3. as a Word document. Same information as above in a Word file.

Steps to setup the Questionnaire

1. Create a space with the example AI assistant config file

   1. Adjust the scopeIds parameter if the questions should be answered on a limited number of documents

2. Activate the Upload in Chat button

3. Add suitable Suggested Prompts:

   Open

   

4. Publish space

Steps to run the Questionnaire

1. Upload an excel that follows the same structure as the example file below

2. Trigger the questionnaire with a prompt: Answer all questions from the uploaded file

3. Once all questions are answered, the generated word and excel file can be downloaded

Required and optional modules

The following modules are required/optional for this assistant:

Example AI Assistant Configuration

Download the provided TXT file and upload it into a new space as an AI Assistant configuration.

The parameter scopeIds needs to be adjusted if the questions should be answered on a limited number of document in the knowledge center. This parameter is a list of scopes ids, e.g. scopeIds = ["scope_id_1", "scope_id_2"]

Example Excel File

Upload the word file into the space and trigger the extraction with the prompt:

• “Answer all questions in the questionnaire”

Open