Abstract [eng] |
The system of legal documents of the Republic of Lithuania is structured though very broad. When analyzing legal documents and in order to identify the basic terms of the law in the articles, it encounters manual annotation, which requires a large investment in time. After a semantic automatic annotation analysis of existing solutions and analysis of legal documents, a new solution was implemented - an automatic semantic annotation system of legal documents. The goal of the new solution is to optimize and facilitate the work of people in the field of law. The program annotates the codes in the article, highlighting the basic concepts of law, thus facilitating the recognition of the terms for the user. The solution implements the interface with semantika.lt. The results of the test, the experiment and the criteria of sensitivity and precision are calculated, which shows that the system recognizes the terms of the norm in the article correctly. For the future system improvements, it is advised to extend the rules of annotations. |