Web services for predictive text analytics

September 16, 2017

Swahili added to Textgain's text analytics services

Two new web services have been added! You can now process Swahili content using Textgain's text analytics services.
I spent a large portion of my academic career working on language technology for African languages. Me and my colleagues, Peter Waiganjo Wagacha (University of Nairobi, Kenya) and Gilles-Maurice de Schryver (Ghent University, Belgium) were among the first in the world to apply machine learning techniques to subSaharan African languages, such as Swahili, Gĩkũyũ, Zulu and Northern-Sotho.

It was fun to return to this topic and develop some Textgain-tools for Swahili, a language spoken by over 50 million people in Tanzania, Kenya, Burundi and many other countries in Central and East Africa. We are very excited to roll out two new classifiers for Swahili: part-of-speech tagging and sentiment analysis. Africa's Voices will be the first to use these services, but they are of course available to all of our customers.

This was a great exercise in adding a web service for an underresourced language. We are sure there will be many more to come.

Guy