Learn About Patents Around the World

 
 
 
IN THIS SECTION
3 posts
RANDom Student
Last seen: 10/09/2023 - 10:54
Joined: 03/09/2023 - 21:19
Filtering granted patents by CPC code

Dear PatentsView Community:

I would like to filter the approximately 7.7M granted patents in g_patent to specific CPC codes. After filtering g_patent to utility patents, which are the patents of interest to me, I think I need to join CPC codes from g_cpc_at_issue, which contains approximately 20.0M rows, to patents from g_patent to filter by CPC code. I have three questions to help:

  1. Can you provide a brief description of g_cpc_at_issue? The table description column on PatentsView's Data Download Tables page is blank for g_cpc_at_issue. Generally, would CPC codes in g_cpc_at_issue be appropriate for filtering?
  2. The name g_cpc_at_issue suggests that the CPC codes in this table are the codes given to a patent at the time it was issued and might not be the codes that would be given to a patent under the current CPC system. The name of another table on the Data Download Tables page, g_cpc_current, suggests that this table might have codes given to a patent under the current CPC system, although it contains a relatively small number of rows (<250K) relative to g_patent. Can you suggest how (if at all) g_cpc_current should be used in combination with g_cpc_at_issue to attach current CPC codes to granted patents from g_patent?
  3. g_cpc_at_issue contains a variable called cpc_type, which is populated with 'additional' and 'inventional'. This post says that 'inventional' means a CPC code is directly related to an invention and 'additional' means a CPC code is not directly related to an invention. g_cpc_at_issue also contains a variable called cpc_sequence. This post indicates that cpc_sequence == 1 means that a CPC code is the primary classification for a patent. Thus, if I would like to filter patents by the CPC code that best represents the subject matter of the patent, then I think I would need to use CPC codes where cpc_type == 'inventional' and cpc_sequence == 1. Can you say whether this is correct?

Thanks in advance for your help, and for making PatentsView an amazing resource!

Sun, 10/08/2023 - 11:39
Russ
Last seen: 03/21/2024 - 09:05
Joined: 11/14/2017 - 22:15
probably a bad data load

Hi RANDom Student,

The CPC changes over time are listed on https://www.cooperativepatentclassification.org/cpcSchemeAndDefinitions As mentioned in two other posts (https://patentsview.org/forum/7/topic/748 and https://patentsview.org/forum/7/topic/756) the current CPCs are probably wrong since the last data load.  The patent office releases the current CPCs of utility patents in bulk each month, see https://bulkdata.uspto.gov/data/patent/classification/cpc/  I believe that is the patentsview source of the current CPC field though that's not what's currently reflected in the database or in g_cpc_at_issue.  There's a US_Grant_CPC_MCF.docx in the bulk CPC directory.  It explains a "U.S. Patent Classification Type Code" which is probably the source of the cpc_type field.  On the sequence, I believe it's zero based, not one based in the patentsview data.  In the data dictionary for g_cpc_current the example cpc_sequence is zero. I would guess that all the CPCs with cpc_sequence == 0 would have a cpc_category of "inventional", but I've only checked at a handful of patents.  There's more on cpc_sequence here https://patentsview.org/forum/24/topic/673#comment-624

Depending on what you are doing, it might not make sense to look at the at-issue CPCs of all utility patents as a whole.  They wouldn't have all been classified under the same scheme.  Use the bulk uspto source of their current CPCs or see what pvteam replies about the current data load.

Russ Allen

Wed, 10/25/2023 - 16:02
PVTeam
Role: moderator
Last seen: 04/26/2024 - 10:39
Joined: 10/17/2017 - 10:47
CPC Data Correction

Hi, thanks for your patience. Russ’s response was accurate.

As you found previously, the g_cpc_current table was missing a significant amount of rows. It was incorrectly truncated during our export process. A corrected file has been made available on the Data Download Tables page. We have also added a table description for g_cpc_at_issue.

Let us know if you have any further questions,

PVTeam