Using Chatterbox NLP Framework to create Smart Data Driven Applications

At Chatterbox Labs we have been working with many partners across the world who have use cases whereby through the application of advanced machine learning new solutions to real world problems could be solved.

These use cases are wide ranging both in terms of industry and specific functions; however they all have one thing in common – data; and lots of it!! So universally the need to make sense of data; to provide understanding to it and classify and categorise it at scale and speed is required.

Our NLP Framework is used by partners to create products in days; literally a few days with minimal training or setup. Use cases are identified, data is ingested; our NLP Framework has all of the processes and methods of 7+ years of research and development to allow partners to manage data training and annotation, to test and evaluate before deploying in a real time environment. The deployed product delivers classification of data in real-time and problems expected to take months to address are solved in days.

Use Case Examples

Customer Helpdesk requests

Improving the efficiency and lowering the cost of managing a helpdesk through the advanced classification of helpdesk requests arriving via many different sources

Customer Feedback on Social Media

Automating the classification of feedback on social media in real-time into complaints, compliments or suggestions

Automating repetitive data intensive statutory tasks

Create machine learning products to produce highly accurate (99%+) classification of millions of data sets linked to financial controlling


Our partners can create machine learning products in days once a specific use case has been identified. They simply take a sample of data (we recommend at least 1000+ rows per classifier) and then create a classifier that is then quickly and efficiently taught what it needs to look for. Once the desired accuracy has been achieved through our training process the classifier can be published and deployed. Multiple classifiers can be wrapped into a single product or split into multiple variables of products – all down to the partners requirements the use case they are trying to solve.

No need to code / recode

In simple terms you don’t need to touch code if you don’t want to. Building a new machine learning product from scratch would take many data scientists and developers months to achieve. Our NLP Framework has all of our processes and methods built in; little if any coding required so it takes days not months; far more efficient and cost effective and capable of quickly addressing dynamic market or customer changes.

If you are looking for a NLP Statistical Machine Learning framework to create smart, data driven applications in days please visit our web site for further information or contact our VP Strategic Alliances, Andrew Watson,