Question & Answer
This model provides answers to questions that users ask from the knowledgebase of data that is acquired using Qualetics Data Streaming. The model is an enhanced form of Semantic Search and data ingestion through Qualetics is required for successful execution and results.
In order to return successful results, the data being streamed will need to include the information required for indexing in the Object Data Object as part of the Object Attributes.
The model is ready to return results immediately after successful streaming of the data.
For successful implementation of the Q&A model, usage of the SessionID is important. By using the same sessionid in each execution of the Q&A model, the responses returned will be contextual and related to the previous responses the model provided enabling an intelligent conversation between the AI agent and the user, provided the relevent data exists.
If the relevant data does not exist, the model will respond with a generic response but this can be fine tuned by providing detailed instructions in the response guidance to not only provide the response on a successful match but also guidance to provide the right response when no information is present.
Rest API Input Example
The input query entered by the user.
The topic of the question. This can be dynamically supplied by the app where the user is performing the search
Detailed guidance to the model to provide a contextual response.
For example: You are Qualie, an AI agent designed to help users understand how to use Data Machines based on the information available. If no answer is found, prompt for more detailed information and provide guidance to reach out to [email protected] if the answers are not satisfactory.
Rest API Example
AI generated response based on the input query and search results
Rest API Output Example
Every model execution output consists of the following standard output parameters
- The input string required for the model to extract the categories
- original input
- This is the input provided to the first step in model which is retained across multiple steps in a Data Machine workflow.
- final result
- The result of the model executed in the final step of the Data Machine workflow
- A unique session id that is generated for every execution of a Data Machine which can be used to retain results across multiple sessions
- The result of the Data Machine execution. If all of the steps in a sequence are successfully executed, a value of "Completed" is provided. If the execution is interrupted at any point, a value of "Terminated" is provided with the reason for Termination.