A large part of the information is mediated to us via texts: in shape of news, emails or whichever documents. This information carries a huge opportunity for the business, which frequently does not find appreciation or use.
In order to make the important information work for your benefit, it firstly has to be extracted from the text. STACC has a long-term experience in Estonian language based textual mining, and the tools what we have created make the task considerably faster and more efficient.
Our cooperation with the Estonian E-Health Foundation is a good example of a typical text mining project. In this case we acquired the documents or epicrises collected by Estonian National Health Information System as an input. Once the data was collected and preprocessed for analysis, we could start with information extraction. Our data scientists chose the most adequate algorithms that helped us to find valuable information from the texts and present it afterwards in a proper form for carrying out analyses. Finally interpretation of results and creation of useful knowledge happened. In this case, based on the epicrises of Digital Story we found new opportunities to carry out medical research and develop new health care software.