Home / DialogFlow ES / How to maximize Dialogflow ES accuracy?
DialogFlow ES

How to maximize Dialogflow ES accuracy?

Here are some tips to maximize the accuracy of your Dialogflow ES bot

Use versioning

There isn’t any meaningful way to measure the accuracy of your bot if you keep adding new training phrases and new intents and also measure the accuracy even as your bot is changing. Instead, use versions, and calculate the accuracy for each version of your bot.

Once you have set up some kind of versioning system, try to improve the accuracy and deflection rate for each version over time.

Design your bot using the explicative approach

Even before Dialogflow CX was released, I proposed that Dialogflow ES bot developers use the explicative approach to designing conversation flows.

In hindsight, I see that I was encouraging people to think of the conversation state machine as they designed their bot.

Don’t use the Training feature (blindly)

It is very tempting to map a user utterance using the Training feature. In fact, you can use it in two ways. You can take a correctly mapped user utterance and add it as an additional training phrase into that intent.

Or you can take an incorrectly mapped user utterance and add it to either the correct intent or perhaps create an entirely new intent.

I recommend that you don’t do this impulsively. Adding a user utterance from the Training feature can reduce both the cohesion and the separation of your bot and make it less accurate over the long run.

Instead, give some thought to how you are designing the training phrases inside your intents.

Set up automated tests

This is probably a little tedious, but it can be very useful if you have a lot of training phrases in your bot. Instead of trying to keep everything in your head, set up and run automated tests as you make changes to your bot.

Add buttons where it makes sense

I have a chatbot on my website which will give you tips based on the specific bot you are building.

It asks whether the user is building an ES or a CX bot, and sometimes the users say something which may be the right answer but not the one the bot can use.

In such situations, it makes sense to provide prompt buttons (also called as suggestion chips) to make it easier for the user as well as to make the bot more accurate.

<— End of article —>

This website contains affiliate links. See the disclosure page for more details. 
Check out the Learn Dialogflow ES and Learn Dialogflow CX courses if you would like to learn Dialogflow in depth.
"The magic key I needed as a non-programmer"

The custom payload generator was the magic key I needed (as a non-programmer) to build a good demo with rich responses in DialogFlow Messenger. I've only used it for 30 minutes and am thrilled. I've spent hours trying to figure out some of the intricacies of DialogFlow on my own. Over and over, I kept coming back to Aravind's tutorials available on-line. I trust the other functionalities I learn to use in the app will save me additional time and heartburn.

- Kathleen R
Cofounder, gathrHealth

Similar Posts