-
I have created an Web based application that simulates Perplexity AI. AI System understands, refines, and retrieves answers to user queries in real-time (uses Google and/or Brave search API for this purpose). It features multi-agent processing, where different AI components collaborate to classify intent, reformulate queries (Search Engines Optimized), and generate responses (along with sources and follow-up questions). So, I have Intent Classifier Agent, which classifies the intent as either Fact (that it can answer Example: What is the Capital of India?), Comparison (Tell me about Galaxy 24 and Apple iPhone 15), reasoning, general (It may or may not be able to answer, Tell me about AI it may be able to answer where as what is the current stock price of MSFT it cannot) or needs clarification. Needs clarification will send the request back to user for clarification. (Example interaction: User: Tell me about Tesla? Intent Classifier will tag it as Needs Clarification and responds with a question Are you asking about Tesla the company or the inventor?). |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 1 reply
-
Hi @balaji830, thank you for your question. This is possible to do with the process framework, but I realize that we do not have any demos that show how. I can get a sample out to help with this. Are you using python or dotnet? |
Beta Was this translation helpful? Give feedback.
-
I am also interested in this. dotnet please |
Beta Was this translation helpful? Give feedback.
-
@sophialagerkranspandey, I am not sure why @alliscode reply is selected as an answer. A reply is NOT an answer. He was asking for a clarification, that cannot be an answer? |
Beta Was this translation helpful? Give feedback.
Hi @balaji830, thank you for your question. This is possible to do with the process framework, but I realize that we do not have any demos that show how. I can get a sample out to help with this. Are you using python or dotnet?