Request Classification AI Agent - Which Skill To Use?

AlpUtkuM
Mega Sage

Hi all,

 

I have a requirement to check historical RITMs (for a specific catalog Item) and determine If It's enhancement request and set the state accordingly. I am planning to create AI Agent but I am not sure which skill(s) to be used to work with AI Agent.

 

Thanks for the reply in advance.

2 REPLIES 2

rpriyadarshy
Tera Guru

@AlpUtkuM 

 

What are the Data Points which will tell you or AI Agent That Its an Enhancement Request (As per ur Historical RITMs).

 

Is this Data Point Structured or Unstructured?

 

As per me There are Few Items i will check and then i will decide on the Solution Approach Like Its a Rule Based or This needs AI Agent and LLMs.

 

Regards

RP

YashwanthV18760
Giga Guru
 

Hi @AlpUtkuM ,

For your requirement to analyze historical RITMs (Requested Item records) and determine if they are enhancement requests, then set the state accordingly using an AI Agent, the best approach would be:

  1. Use the Natural Language Classification (NLC) Skill:

    This skill is designed to classify text inputs into predefined categories. You can train it with historical RITM descriptions or related fields labeled as "Enhancement" or other categories. Once trained, the AI Agent can classify new RITMs based on their description or other textual data.

  2. Use the Text Extraction Skill (optional):

    If you need to extract specific information from the RITM description or related fields before classification, this skill can help parse and extract relevant data.

  3. AI Agent Setup:

    • Configure the AI Agent to query historical RITMs for the specific catalog item.

    • Pass the relevant text fields (like short description, description) to the NLC skill for classification.

    • Based on the classification result (e.g., "Enhancement Request"), update the state of the RITM accordingly.


Summary:

  • Primary Skill: Natural Language Classification Skill
  • Optional: Text Extraction Skill (if you need to extract specific details before classification)


This approach leverages the AI Agent’s ability to classify text data and automate state changes based on the classification results.