ESTIMATES AND FORECASTING THE RESULTS OF THE GROUP CLASSIFICATION EXPERTISE USING THE AVERAGE PRINCIPLE

The features and key differences of experts work in the framework of classification expertise that associated with the use of non-numerical assessments are considered. A number of tasks are highlighted in order to be solved by the working group at the early stages of classification expertise preparation. The features of the Customer's requirements are investigated, which cause a significant increase in both the quantity and quality of work assigned to the working group in the process of expertise preparation and associated with necessity to form procedures for early forecasting of expertise reliability results in general. An approach is used for experts’ opinion synergy by using the average principle. The implemented approach is based on the use of machinery probabilistic matrices characterizing the error-free classification of objects by various qualifications experts. The developed experts’ opinions synergy method is differs from the known expert assessments methods by the possibility of taking into account the unambiguous and ambiguous statements of experts, and also allows taking into account the experts competence (significance of opinions). This complex procedures formation approach makes it possible to assess the expected reliability of the classification expertise results at the early stages of training on the basis of knowledge about the reliability and experts qualifications (competence) included in the expert group. The developed method and its implementing procedures should be considered as the basis for creating an automated system for the expertise classification supporting the, i. e. tool for the working group. The using of such an automated tool will allow the working group to take into account all kinds of Customer’s requirements and the restrictions formulated by him at the stage of expertise classification preparation. The proposed approach can be used to solve a wider range of tasks related to the preparation of different expertise and will improve their efficiency.

Authors: E. A. Burkov, P. I. Paderno, Е. A. Tolkacheva, F. E. Sattorov

Direction: Informatics, Computer Technologies And Control

Keywords: Classification expertise, selection of experts, competence, working group, training, requirements, reliability (correctness), probability matrices

---

View full article