Indicators for optimization of legislation and law enforcement, methods of their identification and usage based on big data (experience of computational experiments on the judicial acts on administrative offenses established by the Chapter 18 Of the Code of Administrative Offenses of the Russian Federation)

Abstract

The subject of this article is the research tools and assessment methods with regards to optimization of legislation and law enforcement. The paper reveals the experience of computational experiments on the judicial acts on administrative offenses established by the Chapter 18 of the Code of Administrative Offenses of the Russian Federation. The research employs various computer methods, including knowledge modeling, methods of natural language processing and machine learning, as well as the related within the framework of interdisciplinary paradigm methods of systemic analysis and expert assessment. Computational experiments were conducted on the empirical basis formed out of texts of 50,438 judicial acts. On the example of big data on administrative offenses, the article demonstrates the interdisciplinary (from computer and legal perspectives)  interpreted results in the context of usage and identification of a number of indicators for optimization of legislation and law enforcement, primarily – time indicator, indicator of individualization of punishment, and indicator of subject uniformity. The conclusions and generalizations are made pertaining to legislation and law enforcement in this area under consideration. Computational methods and the set of indicators can be the groundwork for making decisions in law policy. The advantages of the proposed methodology consist in objectivity of the conclusions that based on methodology open to public verification, as well as big legal data that ensures accuracy of research.