v3 | Patents
The Enriched Citation API provides the IP5 and the public with greater insight into the patent evaluation process. It allows users to quickly view information about which references, or prior art, were cited in specific patent application office actions, including: bibliographic information of the reference, the claims that the prior art was cited against, and the relevant sections that the examiner relied upon. The API allows for daily refresh and retrieval of enrich citation data from Office Actions mailed from October 1, 2017 to 30 days prior to the current date.
The Enriched Citation API provides the IP5 and the public with greater insight into the patent evaluation process. This API release marks one of the first production implementations of the usage of artificial intelligence (AI) for data extraction at the USPTO. This quick access to an organized and prioritized citation list helps users identify relevant prior art, ultimately improving the accuracy and consistency of the patent examination. It serves as a foundation for future enhancements to the patent examiner search system.
Empowered by state-of-the-art machine learning, AI, and natural language processing algorithms, the enriched citation API analyzes the structure and content of office actions. The API then uses sophisticated information extraction and entity extraction algorithms to accurately locate:
- the statutes used by examiners,
- the claims rejected based on prior art,
- the particular prior art references cited, and
- specific relevant sections in the cited prior art references used.
An entity resolution algorithm is then used to consolidate the extraction results with the reference lists in the application, including references cited by applicants and by examiners. The process is fully automated and highly tunable, and is able to incorporate feedback from manual reviewers to improve efficiency and accuracy.