Skip to content

Latest commit

 

History

History
48 lines (27 loc) · 2.36 KB

conversation-analysis.md

File metadata and controls

48 lines (27 loc) · 2.36 KB
Raw

Conversation Analysis

Why it's useful?

As your copilot gains momentum, you will need to manage and monitor conversations. You can use Gooey.AI's LLM script for conversation analysis.

We can determine when the bot is answering and categorize answers based on subject, gender, user location, and other criteria that are important to your organization.

How does it work in Ulangizi AI?

  1. Here you can see the Overall details of the platform, users, and connected number on the left
  2. On the right you can see basic analytics of "Daily Messages Sent"

  1. You can see the "Daily Usage Trends", "Daily Performance Metrics" and "Daily Feedback Distribution"

  1. Finally, you can go through all the tabs with line-by-line details about the various user conversations, we are also updating the categories section soon.

How to make it?

Key Concept:

  1. This runs after we generate an answer.

https://gooey.ai/compare-large-language-models/?example_id=lbjnoem7\

What you need to do:

  1. Save your AI Copilot
  2. Tweak analysis LLM script using the example here
  3. Start from the answered scripts
  4. Optionally add categorization (later)
  5. Save this LLM script
  6. Go to the Copilot's integration tab to the integration where we need to add the analysis script. Open the settings and add the Saved URL for the Analysis script in the "Analysis Run URL".