Optimization of a Search Information System in a Database of Directive Documents

Abstract

To set optimization tasks in an electronic directive document management system, an enterprise can be represented as a set of functional units performing specific functions. In large enterprises, there is a constant exchange between individual divisions of the enterprise and the documentation bases of the divisions. Solutions to several problems related to document databases are analyzed. Here, the primary concern is the selection of a model that can define the relationship between the query execution time and document volumes. The model of this relationship is close to a linear dependence, which is used when assessing the benefits of dividing the database into natural groups. As a method for optimizing the work of the search information system in the database of directive documents, the partitioning of the database into groups according to the sources of document creation is given. Some advantages of such a grouping are explored. To solve the problem, an experimental Table 1 is provided showing the relationship between the volume of data and the execution time of queries. Ordinary least squares fitting is used to select the linear or hyperbolic models used in this Table 1. It is known that databases of directive documents store documents of local or international significance that are created and recommended for implementation by local or international organizations. To improve performance, documents in a database can be grouped by their importance, by the date they are entered into the system, and by the organization that creates the document. A method for grouping directive documents according to the sources of their creation is presented, as well as some numerical characteristics indicating the advantages of this grouping.