M.S. Anferova1, A.M. Belevtsev2

1,2 Moscow Aviation Institute (National Research University) (Moscow, Russia)

1 gludkina@yandex.ru, 2 ambelevtsev@yandex.ru

The current stage of scientific research and technology development is characterized by high dynamics of the formation of new directions and technological trends, as well as the trajectories of their development.

Along with this, the intensification of global competition puts forward as a priority the task of ensuring technological sovereignty and technological superiority based on the development and implementation of strategies for scientific and technological development at the national and corporate levels.

In turn, the development of effective strategies is impossible without solving the problem of analyzing and forming technological trends, studying their characteristics and dynamics of development.

This task can be presented in the following form:

for a given subject area, identify trends and directions in the development of scientific research and technology based on monitoring and analysis of big data in general and special purpose networks.

The known methods and methods of solving the presented problem have several disadvantages:

methodological and technological results have not been found that make it possible to create models and methods for the holistic perception of heterogeneous information by a computer system;

the time of the analyzed publication is not taken into account, when forming trends, the data may be outdated or not identified in the trend due to the fragmentation of information over time;

processing a large amount of information requires a lot of computing power and takes a lot of time.

In the presented work, an analysis of methods and methods for identifying trends in scientific and technological development based on monitoring and analysis of information in general and special purpose networks is carried out.

The research is based on a new approach to the formation of trends and their subsequent analysis based on the proposed clustering procedure with a time offset.

The presented technique provides a solution to the problem associated with the formation of technological trends:

Firstly, the use of a given time interval allows:

Reduce the amount of data processed;

Improve computing efficiency; reduce the processing time of the request.

Secondly, splitting the dataset and entering a time offset allows you to:

Track trends over time;

Provides information about the emergence, evolution and decline of the trend.

Third, the use of top2vec models and graphical representations:

Makes it easier to identify trends;

Visualizes trends in an interpretable way.

Anferova M.S., Belevtsev A.M. The methodology of forming technological trends based on the processing of heterogeneous data in general purpose networks. Science Intensive Technologies. 2025. V. 26. № 1. P. 14−23. DOI: https://doi.org/ 10.18127/ j19998465-202501-02 (in Russian)

1. Anferova M.S., Belevcev A.M., Belevcev A.A. Metodika analiza razvitiya zarozhdayushchihsya tekhnologij i tekhnologicheskih frontov. Izv. YuFU. Ser.: Tekhnicheskie nauki. 2023. № 4. S. 87–96 (in Russian).
2. Belevcev A.M., Sadreev F.G., Belevcev A.A., Balyberdin V.A. Razrabotka intellektual'nyh servisov monitoringa tekhnologicheskih trendov v informacionno-analiticheskih kompleksah. Naukoemkie tekhnologii. 2019. T. 20. № 3. S. 24–29 (in Russian).
3. Anferova M.S., Belevcev A.M. Analiz napravlenij sozdaniya algoritmov effektivnogo poiska informacii v setyah obshchego i special'nogo naznacheniya. Materialy III Vseros. nauch.-tekhn. konf. «Aktual'nye problemy sovremennoj nauki i proizvodstva». Ryazan': RGRTU, 2018 (in Russian).
4. Anferova M.S., Belevcev A.M. Obshchaya koncepciya sozdaniya tekhnologii intellektual'nogo poiska informacii v setyah obshchego i special'nogo naznacheniya. XXV Vseros. nauch.-tekhn. konf. s mezhdunarodnym uchastiem im. prof. O.N. P'yavchenko «Komp'yuternye i informacionnye tekhnologii v nauke, inzhenerii i upravlenii» «KomTekh-2021» (in Russian).
5. Che N.C. CiteSpace II: Detecting and visualizing emerging trends and transient patterns in scientific literature. Journal of the American Society for In-formation Science and Technology. 2006. V. 57(3). P. 359–377.
6. Anferova M.S., Belevcev A.M. Poiskovye roboty dlya avtomatizirovannogo monitoringa informacii v setyah obshchego i special'nogo naznacheniya. 18-ya Mezhdunar. nauch.-prakt. konf. «Upravlenie kachestvom» 2019 g. (in Russian).
7. Mazov N.A., Gureev V.N., Glinskih V.N. Prioritetnye nauchnye napravleniya s pozicii bibliometricheskih issledovanij. Trudy GPNTB SO RAN. 2021. № 1. S. 89–94 (in Russian).
8. Anferova M.S., Belevcev A.M. Analiz napravlenij razvitiya tekhnologij monitoringa v usloviyah bol'shogo ob"ema nestrukturirovannoj informacii. XXIV Vseros. nauch.-tekhn. konf. s mezhdunar. uchastiem im. prof. O.N. P'yavchenko «Komp'yuternye i informacionnye tekhnologii v nauke, inzhenerii i upravlenii» «KomTekh-2020» (in Russian).
9. Mikolov T. et al. “Efficient Estimation of Word Representations in Vector Space”. In: International Conference on Learning Representations. 2013.
10. Anferova M.S., Belevcev A.M. Razrabotka algoritmov intellektual'nogo servisa poiska i monitoringa informacii. Izv. YuFU. Ser.: Tekhnicheskie nauki. 2021. № 3. S 6–17 (in Russian).
11. Anferova M.S., Belevcev A.M. Analiz trebovanij i razrabotka algoritmov intellektual'nyh servisov monitoringa. Izv. YuFU. Ser.: Tekhnicheskie nauki. 2022. № 3. S. 119–129 (in Russian).
12. Anferova M.S., Belevcev A.M. Razrabotka algoritma opredeleniya opornyh tem dlya resheniya zadach strategicheskogo analiza. Sbornik dokladov. Materialy Vseros. nauch.-tekhn. konf. s mezhdunarodnym uchastiem «Komp'yuternye i informacionnye tekhnologii v nauke, inzhenerii i upravlenii» «KOMTEKH- 2022». 2022. S. 132–138 (in Russian).
13. Belevcev A.A, Belevcev A.M, Balyberdin V.A. Metodika analiza i ocenki prioritetov tekhnologicheskih trendov i tekhnologij. Izv. YuFU. 2022. № 6 (in Russian).
14. Sahar Behpour et al. Automatic trend detection: Time-biased document clustering. In: Knowledge-Based Systems 2021. S. 220
15. Deng Z., Liu R., Xu P., Choi K., Zhang W., Tian X., Zhang T., Liang L., Qin B., Wang S. Multi-view clustering with the cooperation of visible and hidden views. IEEE Trans. Knowl. Data Eng. 2020.
16. Aziz S., Dowling M.M., Hammami H., Piepenbrink A. Machine learning in finance: A topic modeling approach. SSRN Electron. J. 2019. http: //dx.doi.org/10.2139/ssrn.3327277.
17. Tan P.-N., Shtejnbah M., Kumar V. Vvedenie v intellektual'nyj analiz dannyh, Eddison–Uesli. 2005. glava 8 (in Russian).
18. Belevcev A.M., Balyberdin V.A., Belevcev A.A. Metodika ocenki vremeni i stoimosti realizacii tekhnologicheskih trendov v usloviyah neopredelennosti i ne polnoty informacii. Naukoemkie tekhnologii. 2019. T. 20. № 5. S. 37–44 (in Russian).
19. Belevcev A.M., Balyberdin V.A., Belevcev A.A., Markelov E.B. Nekotorye tendencii razvitiya informacionnyh tekhnologij dlya sistem setecentricheskogo upravleniya. Vooruzhenie i ekonomika. 2020. № 1 (51). S. 21–26 (in Russian).