The system is based on artificial intelligence and compares scientists

[tanjimajuha20]

Scientists from the Perm National Research Polytechnic University (PNRPU) have developed a unique system that analyzes the digital footprint of Russian scientists (their publications). It will help find a suitable "pair" among scientists and tell you how to quickly contact them.

Polytechnicians have found out finland whatsapp number database that one of the obstacles to scientific activity is finding suitable fellow scientists for joint work. To determine the area of ​​interest of a scientist, it is necessary to study his publications in detail, and this requires a lot of time and deep immersion in the research topic. Contacting the person of interest is also not easy: many researchers work in close teams for years and rarely go beyond them, do not leave direct contacts (publications usually contain only the correspondent address of one of the co-authors). Then PNRPU scientists decided to create a system for interaction between scientists, students and other interested parties. It received the format of a web application - a program that opens directly in the browser.

with each other based on the degree of similarity of their scientific activities. To do this, it analyzes the digital footprint of scientists, namely their publications (articles, books, conference materials), since this source of information is well structured and contains a lot of additional data (co-authors, keywords, etc.).

"By comparing the digital footprint of two scientists, the system calculates and determines the degree of their similarity, receiving a value from 0 to 1. If the value is close to 0, then the scientists' activities do not intersect in any way, and if it is close to 1, they are engaged in similar research," explains Evgeny Kononov, a second-year master's student at the Faculty of Applied Mathematics and Mechanics at PNRPU.

The user can search for scientists by their full name or their identifier in Scopus (a unified database of scientific literature). In addition, a search is available by a short text description (keywords, a fragment of the article title or abstract, etc.) and a search for the closest colleagues of the selected scientist.

The system will not only provide information about scientists, but also suggest how to contact them faster and easier. To do this, it will build a social graph - a set of connections between scientists, presented as a geometric figure of dots and lines. It helps analyze the chain of mutual acquaintances between the user and the scientist he needs and find common ground, increasing the likelihood of acquaintance for the purpose of further cooperation. The system's database includes more than 500 universities and scientific organizations (for example, divisions of the Russian Academy of Sciences) of the Russian Federation.

"We plan to distribute the system in two versions: open and closed. It will be available directly through the browser; authorization will be required to access paid functions. The closed version will have expanded functionality (for example, the ability to build a map of the competencies of scientists and educational institutions), and will allow for a more qualitative search due to a flexible system of filters (by keywords, by the date of the last publication)," concludes Vadim Danelyan, a postgraduate student in the Department of Computational Mathematics, Mechanics and Biomechanics, and the head of the youth design and technology bureau group of the advanced engineering school "Higher School of Aviation Engine Building" of PNRPU.

As scientists from Perm Polytechnic University note, there are no analogues of this development in Russia, and foreign tools usually offer the user only articles of interest or upcoming conferences. The developed system will help quickly and easily find specialists to form strong scientific teams and work together on large-scale projects, this will increase Russia's scientific sovereignty.

Polytechnicians plan to add scientists and scientific organizations from other countries to the database. In addition, there are plans to improve the system: improve the search, increase the number of filters and functions, for example, adding a personalized mailing list about upcoming grants, competitions, seminars, etc. A full release of the system is planned for the end of 2024.