Learn About Patents Around the World

 
 
 
IN THIS SECTION
2 posts
Yishu
Last seen: 10/02/2023 - 23:44
Joined: 09/29/2023 - 03:37
Clarification on Pre-Grant Dataset and Grant Dataset

I am seeking clarification on an inconsistency I have encountered within the bulk download datasets.

According to the PatentsView Data Logic Diagram, the ‘pg’ data comprises ‘pre-granted data only.’ This implies that once a patent application is granted, it should no longer be part of the Pre-Grant Datasets, such as ‘pg_published_application.’

However, my analysis reveals instances where applications listed in ‘pg_published_application’ have indeed been granted, as evident when cross-referenced with the Grant Dataset ‘g_patent.’ Surprisingly, these records continue to feature in the ‘pg_published_application’ dataset, contrary to expectations.

A plausible explanation could be that granted patents are intentionally retained in the Pre-Grant Dataset. This assumption aligns with the statement in the Pre-Grant Data Download Dictionary, which states

The pre-grant publications data includes all publications released by the USPTO for download from 2001 through the most recent data update.’ Despite this, I’ve noted the absence of numerous granted patents from ‘pg_published_application,

suggesting that some patents are indeed removed from the Pre-Grant Data after being granted.

For your convenience, I have attached a file detailing the patent_ids in question, indicating whether each ID appears in ‘pg_published_application’ or ‘g_patent.’

I am at a loss to reconcile this inconsistency and would greatly appreciate any clarification or additional information that PatentsView could provide on this matter. Ensuring a consistent and accurate understanding of the datasets is crucial for the research community, and an update or clarification in the dictionary would be immensely beneficial.

Thank you very much.

Tue, 10/03/2023 - 18:39
PVTeam
Role: moderator
Last seen: 04/23/2024 - 10:43
Joined: 10/17/2017 - 10:47
CLARIFICATION ON PRE-GRANT DATASET AND GRANT DATASET

Hello Yishu

Happy to provide some clarification.
After an application for a patent is submitted, it may undergo some modifications to its content and/or interested parties before it is ultimately accepted or denied. Additionally, documentation of the examination itself is added to the patent document which would not be possible to include in the original application. As a result, the publication of the application and the publication of the patent are always considered separate documents, and the data concerning them are likewise maintained separately, regardless of whether an application is ultimately successful in receiving a patent.

Furthermore, in contrast with patents, which are always made public, applications may be excepted from publication under certain circumstances, including withdrawal from consideration during the standard publication delay of 18 months and submission of a valid non-publication request from the applicants. You can find full details of the publication and non-publication criteria on the USPTO website here. Because of this and because of differences between the examination time and the standard 18-month publication delay for applications, it is expected that there will be some patents that will not have a corresponding published application and that there will be some published applications which will not have a corresponding patent.

Our table pg_granted_pgpubs_crosswalk shows the correspondence between granted patents and their published applications when such a correspondence exists. When a patent or pgpub does not have a corresponding document of the other type, the crosswalk indicates this as well. 

Maintaining both the pre- and post- examination documents allows for analysis of any changes that are deemed necessary to convert an application into a successful patent and of patterns in assignment of rights during review of an application.  

To address your observations specifically, the logic diagram's indication that some tables pertain to "pre-grant data only" is intended to reflect that a pre-grant table provides the data as shown on an original application document, disregarding any changes that may have been applied during examination. Likewise, a table marked as containing "granted data only" indicates that the data in that table is presented as it appeared on the final publication of an accepted patent regardless of how or whether the corresponding data appeared on a published application. Both of these labels contrast as well with the label "table contains both granted and pre-grant data", which applies only to the crosswalk table that links corresponding patents and pre-grant publications.

If there is any further clarification we can provide, please let us know.

Best,
PVTeam