COMPUTATIONAL MODELS FOR BUSINESS AND ENGINEERING DOMAINS 67 MULTIDIMENSIONAL NETWORKS FOR HETEROGENEOUS DATA MODELING Sergey Maruev, Dmitry Stefanovskyi, Alexander Troussov, John Curry, Alexey Frolov Abstract: Big data frequently come in tabular form of rows and columns of numbers, special codes and short textual descriptions, in strict, structured, disciplined formats generated by a variety of transactional and operational business systems. In this paper we discuss the advantages of modeling heterogeneous data by multidimensional networks in line with the concept known as “Graph databases”. Graph-based methods provide a powerful abstraction for mining such data; however, it is hard to achieve good results in mining using of the shelf methods. In this paper we show how empirical methods of fuzzy logic could be injected into abstract graph-based methods to achieve desirable results. We outline the wide range of applications of that modeling and mining, and present our results on the use of our methods of modeling and mining for processing of custom declarations for commercial goods. We examine several use cases, including recommendations to custom officers and participants of the international trade. The feasibility of the approach was tested by application to 2500 custom records collected during a continuous period of one month at eight border checkpoints between Russian Federation and two EU countries. In several use cases the algorithm achieved high accuracy under experimental conditions. Keywords: big data, graph-based methods, custom declarations. ACM Classification Keywords: Algorithms, Economics, Experimentation, Theory. Introduction Big data frequently come in tabular form of rows and columns of numbers, special codes and short textual descriptions, in strict, structured, disciplined formats generated by a variety of transactional and operational business systems. In this paper we discuss the advantages of modeling heterogeneous data by multidimensional networks in line with the concept known as “Graph databases”. Graph-based methods provide a powerful abstraction for mining such data; however, it is hard to achieve good results in mining using of the shelf methods. In this paper we show how empirical methods of fuzzy logic could be injected into abstract graph-based methods to achieve desirable results. We outline the wide range of applications of that modeling and mining, and present our results on the use of our methods of modeling and mining for important area of applications - processing of custom declarations for commercial goods. International trade is one of the most important drivers of the global economy. Therefore, the study of impediments to this trade is of interest to the field of international economics. International trade is typically more costly than domestic trade due to the imposition of extra direct and indirect costs including tariffs, time costs due to border delays and processing costs that are exacerbated by differences in language, legal system and culture, see, for instance, [Zvetkov et al. 2013] in Russian. We examine several use cases, including recommendations to custom officers and participants of the international trade. The feasibility of the approach was tested by application to 2500 custom records (which have 12043 items of goods) collected during a continuous period of one month at eight border checkpoints between Russian Federation and two EU countries, the same data set that was used in [Maruev et al. 2014].