Разработка подсистемы интеллектуального анализа данных для системы электронного документооборота Citeck

Abstract

An urgent task in the implementation of electronic document management systems (EDMS) is to expand their functionality through personalization and taking into account individual characteristics of the organization. The article deals with expanding the functionality of EDMS by designing a subsystem for data mining. As part of the study, the principles of formalizing the processes of processing incoming correspondence and organizational and administrative documents, methods of collecting and analyzing data on the work of users with various types of documents through the use of artificial neural networks and a comprehensive assessment of improving the efficiency of the EDMS of an educational institution were studied. Quantitative characteristics that directly affect the process of monitoring the execution of orders have been determined. This is the time spent for creation of the document and execution completeness of the document. A mathematical model of the process of creating documents based on data from the EDMS has been developed. Regression coefficients have been calculated. Analytical dependences of the quality of the developed documents on the time of their execution and volume have been obtained. The scientific novelty of the research lies in the development of an algorithm and software for automating the collection and analysis of data through the use of neural networks in the EDMS. The main scientific results include formalized criteria for documents and stages of their development, the algorithm of the mining subsystem, the developed software for the EDMS of the Lyceum. The results obtained made it possible to identify the types of documents and the stages of their development that are most demanding on the resources necessary for their implementation, which can later be used to find ways to optimally organize work on the preparation of documents of various types.