QBox Core

Significantly reduce the time and manual effort it takes to train NLU data models

How does QBox evaluate your chatbot?

What is QBox?

The QBox algorithm is designed to test reliably and effectively where standard techniques such as K-fold are less effective.

Simply connect your model to QBox and it will analyze every utterance to provide you with a meaningful score that you can use to validate your fixes and uncover further performance opportunities.

QBox facilitates different types of tests that you can conduct to suit where you are in your chatbot building journey. So whether you are beginning to identify issues or you are at an advanced stage and subjecting your chatbot model to more challenging test data; QBox will see you through your journey.

See a demo

Using QBox will enable you to:

Gain the confidence to go live

Scale your chatbot domain

Put your subject-matter experts in charge

Measures understanding - It’s not easy knowing when your chatbot has had sufficient training (or even to quantify the quality of this training).

With the QBox scoring system, you can measure understanding, and recognize when your chatbot has reached the maturity level to serve your customers efficiently and effectively.

Use QBox to regain control of your chatbot testing, and track and analyze the incremental progression of understanding of your Artificial Intelligence chatbot as you train it.

Increases knowledge - At the beginning, your chatbot will inevitably cover a smaller subject area than your long-term plan has. But what happens when you expand the domain that your chatbot covers, and increase its knowledge?

QBox allows you to validate your chatbot’s knowledge as you expand it. This helps you rapidly identify whether you’re causing regressions in knowledge areas that were previously performing well.

Easy to use - Your subject-matter experts are best placed to train your chatbot because they know their subjects better than anyone else.

QBox is easy to use; it doesn’t matter if your subject-matter experts have a background in computer/data science or not. Everyone and anyone can monitor and train AI chatbots.

Key features

As well as providing insight into the contents of your chatbot model, QBox will also provide the insight to enable you to keep your chatbot model healthy and in tip-top shape as you continually improve and scale it.

Go further with QBox add-ons

QBox offers several features that extend the standard QBox Express experience; Want to quickly try out model experiments without having to go through the full flow of making trial and error changes to your model? Need deeper analysis to understand why certain utterances are misbehaving? Or are you guessing what your optimal confidence threshold is?

Here's just some of the add-ons that QBox offers:

QBox | Experiment


Make changes to your model from within QBox, and quickly assess the potential for improvement and regression before pushing those changes to your provider – really helpful for evaluating possible additional utterances or fixes.

QBox | Word influence analytics

Word influence analytics

Allows you to see to what degree each word in an utterance influences the prediction. This takes analysis to the most granular level and can be key in understanding the prediction.

QBox | Collaboration


Writing enough training or testing data is tough. QBox can give you jump start by analyzing your training data and suggesting potential gaps to fill.

QBox | Confidence threshold optimisation

Confidence threshold analysis

Helps you assess what is the best confidence threshold to set within your chatbot NLP engine.

See QBox in action

QBox is an excellent tool that provides a comprehensive and customizable application for testing conversational AI applications. It simplifies and accelerates the testing process, improves accuracy, and helps users deliver high-quality digital assistants.

See a demo