The goal of Tsert (teeSert) Inc. is to become one of the major providers of software applications, systems, and services. We intend to provide Natural Language Processing based engines, to scan, understand, and translate unstructured text. Fully automated translations of general, legal, technical, medical, pharmaceutical, scientific and engineering texts, have been made accurate enough to only require the hiring of proof-readers. We also provide the design, development, and testing of applications and products through our R&D team.
Our statistics based algorithm and process, centering on word type disambiguation, forgo inter-linguas, which are usual in translation systems, and cannot be circumvented using either Bayes theorems, formal grammar methodologies, or weighted graphs and multigraphs for establishing the rules set and recognizing sentences.
Look at the state of translation technology by reading what is on the following page. You will find on that page some of what we are doing. The semantic part of our design, dealing with word sense disambiguation, is based on previous work.
Matching the results of our translation system without using our approach is not possible. Test other systems using the 17 files, english to french, we put on our website. Statisticl textual replacement based translation systems, where previously human-translated text is cut and pasted into the document, do not match our system.
Real-world examples of the results of our content engine are on the following page.
The first release of the PI-Desktop©® environment will be in March 21st, 2012.
The first release of the Kodiak.light©® version of our OS will be in April 23st, 2012.
The first release of the Kodiak©® version of our OS will be in July 23st, 2012.
The completion of the Data©® , natural language question & answer, Android©® application will be in July 17th, 2012.