Patent analysis

This article may need to be rewritten to comply with Wikipedia's quality standards, as (for example) the introduction does not appear to properly summarize the body of the article. You can help. The talk page may contain suggestions. (September 2023)

Patent analysis is the process of analyzing the texts of patent disclosures and other information (such as priority dates, filing and issuance countries, patent maintenance payments, patent citations, patent infringement actions etc.) from the patent lifecycle. Patent analysis is used to obtain deeper insights into different technologies and innovation. Other terms are sometimes used as synonyms for patent analytics: patent landscape, patent mapping, or cartography.^{[citation needed]} However, there is no harmonized terminology in different languages, including in French and Spanish. Patent analytics encompasses the analysis of patent data, analysis of the scientific literature, data cleaning, text mining, machine learning, geographic mapping, and data visualisation.^[1]

Patent analytics is used in industry and increasingly^{[as of?]} explored by the public sector to take informed decisions related to prioritization and investments in R&D, IP portfolio management, commercialization of technology, and research collaborations among others.^[2][3]

Patent documents also present a very unique structured and meaningful data collection, which is not a subject to any copyright restrictions (such as data mining prohibition), and for this reason it has received a great attention from data scientists and from economists.

Patent analysis tools and methods have traditionally^[when?] been done using spreadsheet-based data analysis methods, such as SQL. However, since ca. 2020 the field of patent analysis has witnessed a convergence of traditional patent analytics with data science, machine learning, semantic technologies, and artificial intelligence^[4] along with a surge in available tools that are being applied to patent visualization.^[5] There has also been an increase in open-source software, tools^[6] and datasets^[7] being used for patent analytics, as well as the use of techniques, such as machine learning, for different tasks.^[8] Some tools^[9] propose semi-automated production of visualizations, dashboards or reports. Terabytes of patent information from many patent offices is available on-line for free from INPADOC or espacenet or Patentscope. Many developers of big data software, such as Google Patents, The Lens, Clarivate Analytics, ip.com, Derwent World Patents Index, and Questel-Orbit, use these free and other patent databases to test the capabilities of their own data analysis software.^{[citation needed]}