Botium Introduces NLP Analytics to Cover All Aspects of Conversational AI Performance Evaluation

Botium, a market leader company in chatbot testing and training, is thrilled to announce the launch of their NLP performance testing system called Botium Coach.

The Austrian startup is dedicated to address the challenges of testing chatbots. Their new system is based on an open source testing tool, that the community terms as “Selenium for Chatbots”. Botium allows testers to easily automate the testing of chatbot conversations. Tests can be executed in seconds, and they are repeatable with every software update.

On top of this, Botium provides an entire management toolbox with predefined test environments and test sets containing thousands of instantly runnable tests.

With the introduction of Coach they have taken the step on covering all aspects when it comes to chatbot testing and training. Botium is the only company on the market who have been able to combine conversational testing and NLP analytics. Conversational testing enables users to assert the context of the conversation and check all steps have been processed. This form of testing brings you as close as possible to your end user experience.

Botium Coach, on the other hand, provides visual insights to NLP performance metrics, identifies which type of conversation your training data fails to recognise, and suggests steps for improving it. It uses statistical metrics to make NLP model training cycles comparable and points out areas of interest. Botium Coach will detect any issues within test results and recommends actions to improve the overall NLU performance. Or, in other words, it will find weak training data and suggest refinements. For example, by gathering and augmenting training data based on its paraphraser addon.

Find more awesome best practice examples and use cases on the website of Botium.
Botium are happy to hear from our readers directly.