Design of the knowledge base of expert system for management scientific projects

This article describes the process of developing the knowledge base of the expert system for management scientific and innovative projects based on an analysis of the specific situation, the degree of development of the project at any given time, assessment of its condition, as well as the analysis of the previous conditions of work with the project. We present the algorithm and method of bring to the information from the database to the form, which allows to describe the conditions and assess the situation in the form of a rule base for obtainment interim findings and final decisions. On the base of the analysis and formalization of the decision-making process for the management of scientific projects are determined the directions and content analysis of the available data, next, based on the ER model was developed multidimensional data model, allowing to determine the possible dimensions of the cube and their hierarchies. Dataset values are database tables that include quantitative and qualitative data. Formalization and description of the expert knowledge of the studied subject area in the form of conceptual and functional models in the field of knowledge allowed us to determine the conditions and procedure for decision-making and develop a base of rules for the formation of interim findings and final decisions on working with a specific project. With the purpose of detailed presentation of the reasoning ES issued recommendations it was agreed to submit an interim findings and summing the final decision as separate blocks of information. Thus, the final decision in each project depends on a General assessment of the project, a situation in which it is at the moment, as well as conditions of participation in the contest. Block "project Status" contains information obtained from the processing of the calculated estimates of the project, the output status of the project is based on the analysis of the overall assessment, of index of commercial viability and the assessment of completeness. Block "Work with project" outputs guidelines for working with the project, based on the results of the analysis of the situation the project development. Block "Work with application" also lists recommendations for working with the project based on the analysis of previous results of the preparation of applications. Block "Solution" displays a summary of recommendations based on the given conditions, the state of the project and evaluation activity for the project. Thus, the process of obtaining the solutions in ES is a direct chain of inference: the search is based on data (data-driven search), the process of solving the problem begins with the source of the facts, then, by applying valid rules of state change is the transition to new facts, and so until then, until the goal is reached, in this case the results of the recommendations for the specialist for working with the projects.

Keywords

Authors

References