Прогнозирование развития соавторства в написании научных статей научно-технического центра Газпромнефть на основе модели
A collective co-authorship of scientific articles has deterministic and random structural components. In addition to the rational aspects of the team assembling of co-authors of the scientific article there are also emotional components. Over time add up and disintegrate the working groups of researchers, updated the organization workforce and the composition of the contractors that take part in joint industry collaborations for conducting research. Despite the complexity of co-authorship nature, there are several classes of models to simulate the formation of co-authorship. Among them, models based on random graphs and models of the formation of collaborations based on the competencies of the authors. Both the mathematical apparatus developed and used for several decades separately. However, practical applications of the models of collaborations in the corporate practice not so much. The authors put forward the hypothesis that it is necessary to combine several different types of models to have better understand the nature of scientific collaborations in a separate organization. The authors of this study set the task to develop a method of constructing a model of coauthorship for scientific-technical center, including various structural components of joint authorship. As a result, the authors developed a model using machine learning methods, random graphs and models of competencies. Based on the developed model, the forecast for the development of coauthorship in the writing of scientific articles scientific and technical center of GazpromNeft. The practical value of this study is the following: 1. Quantified the contribution of different structural components in forming collaborations when writing scientific articles. 2. Forecasting the development of co-authorship in the writing of scientific articles allows planning enterprise resources to support the growth of scientific publications. Understanding the cluster structure of co-authorship makes it possible to align the lines of scientific activity in accordance with the strategic plan for the development of the scientific and technical center.