Intelligent analytical subsystem of the digital platform of the university
The goal of the practice is to improve the efficiency of the university by developing a system for supporting management decisions in the form of an analytical subsystem of the university's digital platform. The developed analytical subsystem of the digital platform of Bauman Moscow State Technical University is a set of services that provide solutions to university management problems. The main purpose of the proposed practice is to improve the efficiency of making management decisions on the formation of teams of performers, development of scientific areas, use and development of staff, facilities and equipment, scientific and technical reserve of the university through monitoring, intelligent analysis of data and development scenarios, formation of forecasts, conclusions, recommendations and reports. The analytical subsystem functions in the form of a web application and is open for development. Services for training engineering teams for the industry, identifying promising scientific areas, and predicting the emergence of new fields of study have been implemented. The implementation of 11 more services and a mobile application is planned. The proposed solution can be integrated without further development via API into the information systems of other universities. The microservice architecture of the analytical subsystem allows for easy expansion and interaction with third-party resources, as well as the ability to integrate with the unified Russian platform of government services GOVTECH. The implemented service for training engineering teams for the industry based on a technical specification file for R & D or event requirements forms and recommends teams of performers from among employees and students, taking into account their experience. When implementing the system in universities of different countries, it is possible to select international teams. This allows for increasing the efficiency of project activities and the distribution of university resources, as well as implementing the training of well-coordinated teams of graduates for a specific industrial partner. A large BERT language model is used to analyze text documents, and a genetic algorithm based on the punctuated equilibrium method is used to select the optimal group of researchers. The service for determining promising scientific areas based on the analysis of the dynamics of changes in the international patent classifier predicts scientific areas that may appear in the near future, and also generates a forecast for the development of new and existing scientific areas. The service for predicting the emergence of new fields of study based on information about promising scientific areas determines the competencies that professionals must have to conduct research and development in these areas, and generates recommendations for the training of students and researchers in order to develop in-demand competencies.
Artificial Intelligence And Digital Services
Artificial intelligence
1 year
Bauman Moscow State Technical University, Moscow, Russian Federation
The analytical subsystem of the university's digital platform is being developed within the framework of research and development under the Priority-2030 program on the Bauman DeepAnalytics track to solve some problems at Bauman Moscow State Technical University. The analytical subsystem belongs to the class of corporate EIS systems and is designed to manage university processes and improve their efficiency, and also facilitates the integration of efforts of universities in different countries to implement international projects. The analytical subsystem provides a wide range of tools for collecting, processing and visualizing data using advanced artificial intelligence methods. The analytical subsystem has a modular structure and is highly flexible due to the ability to customize each module in accordance with the needs of the decision maker. Unlike the EIS systems presented on the market, this analytical subsystem is supplied with ready-made services implementing decision support for various management tasks of the university, which significantly facilitates its integration into the business processes of the university. When implementing the analytical subsystem in BRICS universities, it is possible to automatically select international teams of performers, including for international projects. The ability to predict scientific areas that may appear in the near future allows the university to adapt and make effective decisions in the modern conditions of rapid technological development. The subsystem allows processing documents in different languages, which facilitates the preparation of proposals for international projects.
The team is the winner of the international university award in the field of AI and big data "Gravity 2024" Media coverage: The system offers a solution // Engineer, 2023, No. 47, pp. 104-105. https://api.www.bmstu.ru/upload/publication/161/651e6e32d1dbf.pdf There are sections of this site dedicated to the product https://da.bmstu.ru. News reports https://bmstu.ru/news/baumancy-oderzhali-pobedu-v-premii-gravitaciya https://vk.com/@-214887108-uchastie-v-vii-vserossiiskoi-pospelovskoi-konferencii-gibrid https://vk.com/@-214887108-baumanskaya-komanda-razrabotchikov-pobeditel-premii-gravit https://vk.com/@-214887108-uchastie-v-kongresse-russkii-inzhener-3-ya-mezhdunarodnaya-k and others.
The results were published in 9 Scopus papers and 14 papers in HAC Russian journals, 28 reports were made at conferences, 4 certificates of state registration of software were received, 1 PhD thesis was defended. Some publications: Berezkin D., Murashov M., Liashenko N. Automated formation of university R&D teams based on the competence selection algorithm // Procedia Computer Science, 2024, V.234, pp.373–380. https://doi.org/10.1016/j.procs.2024.03.017 Berezkin D., Kozlov I., Martynyuk P. Predictive analytics of scientific and technological trends for decision making in university management // Procedia Computer Science, 2024. V.234. pp.270-277. https://doi.org/10.1016/j.procs.2024.03.001
