The Conversational Data tab stores questions collected from user interactions that the Bot did not manage to match any intent to.
How is this useful?
After the bot is live, it is important to view the data collected from questions users ask to continuously train the language model. Adding user questions for training not only improves the Bot’s current vocabulary, it expands it as well so that the Bot can answer more questions posted by users relevantly in the real world.
How does this work?
The Bot automatically picks out real user data and parks it under the most relevant intent. But for data that doesn’t match with any intent, the Bot parks them under the “Conversational Data Tab”
For every piece of data in this tab, you can decide whether to accept it as an example for the current intent, archive it, classify it under a different intent, or create a new intent altogether.
Steps to improve the Bot with real user data
- From the top of the FAQ dashboard, navigate to the “Conversational Data” tab
- On the right panel, you can view all the “Unclassified Messages” – these are all questions collected from user interactions. This includes both questions that the bot is able to answer, and those that the bot is unable to answer.
- On the second column “Prediction” – these are predictions made by the bot. If it is empty, it means that the bot does not recognize the question.
- In the “Confidence” column, it shows how confident the Bot is about the example and intent match it had done, to the max limit of 1.0. The higher the confidence level, the better.
From here there are four actions you can take: