Modeling the processes of inventive activities in innovation ecosystems

The article discusses the urgent problem of modeling the processes of inventive activities under the influence of many factors of uncertainty characteristic for the interaction and mutual dependence of innovation ecosystems subjects. The indicated models of the processes of inventive activities, the interdependence and interaction of the subjects of innovative ecosystems, firstly, are necessary for the understanding and adequate perception of the processes occurring in the ecosystems themselves. Secondly, such models are needed to develop and create intellectual systems for supporting solutions in the field of management and financing of companies and organizations included in such ecosystems, as well as in the field of rights to IP, and for solving other problems. A young innovative company – startup – is considered as the central link of the ecosystem, whose intellectual activity result is ideas formalized in the form of applications for inventions sent to the patent office. Streams of random events of filing patent applications by a startup and receiving decisions from the patent office are considered within the framework of classical probability theory. The ecosystem model, built as a Bayesian belief network, makes it possible to take into account not only the events of patent signals on the emergence of new technologies (inventions) at a startup, but also the cause-and-effect relationships between all subjects of the innovation ecosystem, including a startup, an independent expert technology broker, an institutional investor (venture fund), and the patent office. As a result of training the constructed model and subsequent analytical and diagnostic conclusions, the venture fund is able to extract new knowledge and reasonably reduce uncertainty when making decisions about financing a startup, taking into account the current state of its patent portfolio. The Bayesian model of an innovative ecosystem has been proposed and implemented, which, in fact, is a prototype of an intellectual system for supporting the solutions of an institutional investor (venture fund) on financing (continuation of financing) of a startup. The model takes into account not only the presence of the so-called market patent signal about the emergence of a new technology (invention), but also the judgments of experts of different levels about the quality and novelty of this technology. The ability to quickly retrain the model, for example, in accordance with new knowledge obtained