Objective - While managing the innovation activity, it is necessary to base it on reliable sources of scientific and technical information, including patent research. However, the existing variety and scale of patent databases necessitate the development of an instrument that enables processing large volumes of patent information within limited timeframes. In these conditions, it is necessary to use machine learning (ML) technology to create a solid information base for management decisions.

Methodology - The objective of the study presented in the paper was to propose an algorithm for processing patent data to improve the quality of patent research. The essence of the algorithm is that all necessary patents are ranked according to a relevance criterion, after which the researcher analyzes the already essential patents.

Findings - The paper envisages the algorithm's practical realization using a gravity-driven power generator case. Findings indicate that the proposed new instrument enables a significant reduction in processing time for patent data.

Novelty - The paper contributes to innovation management by integrating patent analytics and machine learning.

Type of Paper - Empirical

Keywords: Innovation activity; patent analytics; machine learning technology; a gravity-driven power generator.

JEL Classification: D80, D81

URI: https://gatrenterprise.com/GATRJournals/JBER/vol10.3_5.html

DOI: https://doi.org/10.35609/jber.2025.10.3(5)

Pages 131–142