Interview: Meet TDIP – Telematics Data Interoperability Platform

Electric cars, telematics and data portability: these are the keywords that drive project TDIP – Telematics Data Interoperability Platform. Curious? Team member Balázs Szabó explains us more in this interview.

Can you briefly introduce yourself?

I am Balazs, CEO and Co-founder of Konetik. Konetik helps businesses to integrate electric vehicles with a software that analyses real vehicle usage data and recommend the suitable electric vehicles and the optimal charging infrastructure.

What is your motivation to work in the data portability field?

While working with telematics companies we realised that there is a lot of value hidden in the second-life use of telematics data. However, data portability is an issue due to the different formats.

How did you hear about DAPSI and what drove you to apply?

We have seen this opportunity in a newsletter and immediately thought that the DAPSI program could be helpful to overcome our data portability challenge: makes it easier for us to integrate with partners and it can be also useful for other companies and the broader community.

In simple words, what challenge does your project address?

Telematics data is notoriously varied in their format. Despite the existence of standards, telematics data providers don’t seem to be actually using them in practice.
Instead of trying to convince data providers to stick to one single standard, our goal is to improve the interoperability of telematics data, so that users and data providers can share data more easily between services, irrelevant of the format they use.

What solution are you developing?

The solution is to learn to map different formats and translate data from one format to another seamlessly. This may seem like a very difficult task to automate. However, recent advances in machine learning, and in particular natural language processing, give us the tools to do so with confidence. We use pre-trained deep neural networks to perform semantic analysis, and perform numerical tests to quantify the similarities between data columns in different formats.
Moreover, using OAuth2, we benefit both final users and data providers with the power to connect their data, in whatever format they want, to whichever services they need, using our open-source API.

What will be the next steps?

The next step of the project is to Integrate with the data providers, calibrate the system to make it even more automated and get ready with a reference implementation.