#natural language #Neural nets

TEXTA OÜ is the first language technology startup in Estonia focused on natural language processing and its application. The main activity of the company, which grew out of STACC in 2016, is the development of open source software for text analysis. The Toolkit developed by TEXTA is the first artificial intelligence software – a web-based text mining framework that was also added to a national code repository.

The period of the project: September 2017 – January 2018

What did we solve?

The TEXTA Toolkit needed to develop and improve a number of functionalities and tools for multilingual data processing. This could be solved in cooperation with STACC.

How did we solve it?

We used machine learning and technologies based on neural networks.

What were the benefits?

The TEXTA Toolkit is now more functional than before and that makes it more attractive to the customers. As a result of research carried out in the framework of a collaborative project between STACC and TEXTA, the TEXTA Toolkit enables the use of multilingual word inputs and deep neural classification models in its functions.

“Cooperation with STACC was very beneficial by virtue of the high level of specialists working there. Of course, I also have to mention the background structures that strongly contributed to the smoothest possible process.” 

Raul Sirel, co-founder of TEXTA