Intellectual Systems and Technologies

Extraction of business process models from texts

K. V. Sokolov^a, D. A. Timofeev^b, A. V. Samochadin<sup>b

a</sup> Saint Petersburg State University
^b Peter the Great Saint-Petersburg Polytechnic University

Abstract: Formal models of business processes are necessary for process analysis and for developing process-oriented services. Process modeling is a time-consuming task which should be performed by an expert. A large part of the expert's work consists of gathering information about the process. In most cases, process descriptions already exist in textual form. The automated method of extracting process models from texts is required to put this knowledge into use. In this paper we develop a process extraction method for texts like records, reports, or user stories. The main feature of the proposed method is the application of process mining techniques to text analysis.

Keywords: business process, process extraction, process mining, semantic analysis, natural language processing.

UDC: 004

DOI: 10.5862/JCSTCS.229.7