Consortium of COMARCH POLSKA S.A. from Krakow and Neurosoft Sp. z o. o. from Wroclaw on 21st of November 2014 signed an agreement to implement a judgments publication system for common courts. Thanks to advanced techniques for analysis of grammatical texts, the implementation of the system will significantly accelerate the process of so-called anonymization (removing personal data from the judgments) which will allow for almost immediate publication of judgments by common courts throughout Poland.

Universal access of citizens to court judgments has a significant impact on the quality of established law. Less than two years ago publication of the current judgment encountered a barrier in the form of manual removal of sensitive data from the texts – the process was laborious and burdened with a high risk of errors, in addition, the manually anonymization result was inconsistent. Thanks to the cooperation of scientific institutions and Neurosoft, we were able to create a system for automatic removal of sensitive content, and the only task of a man is to verify computer -implemented changes.

The project financed by the Ministry of Justice concerns the implementation of the system in 182 courts in which such a solution is not yet operating. As a result, by the end of January 2015 the system will cover all common courts in Poland. Currently, this type of tool works in more than 120 courts and contains the full content of more than 80 thousand judgments generally available on the Internet in one place, in the so-called Court Judgments Portal. For over a year, this data base was used by many people; each month the portal is visited by over 270 thousand different users generating over 1 million views.

The construction of such an advanced system would not be possible without the close cooperation between business and science. – Says Cezary Dołęga, Chief Technology Officer.

With the support of researchers from the University of Wroclaw (Faculty of Law, Administration and Economics and the Faculty of Mathematics and Computer Science), we succeeded to analyze a number of practices used by the courts in terms of anonymization what allowed us to create an extensive set of anonymization rules. Then these rules were “translated” into the computer language and the result is a specialized program automatically dispensing the rules for any text of the judgment. Anonymization is not the only unique feature of the developed system. The program also finds references to legal documents, finds thematic entries in the analyzed texts or can automatically find similar judgment- a judgment related to the model judgment.