What happens when your silver bullet isn’t in the abstract?

[tweetmeme source=”Intellogist” only_single=false]

We English-speaking searchers often have to conduct global patent investigations.  In order to search non-English collections, many of us need to rely primarily on English-language abstract files such as the Patent Abstracts of Japan collection.   Another good option is to search the Derwent World Patents Index, where human editors translate and summarize the document into an English abstract, helping to standardize the technical terminology used in the patent.  But as we all know, it’s impossible to shrink all the useful content of a patent into an abstract, no matter how well it’s translated.

For important prior art searches, companies should consider initiating native-language full text searches in collections of particular interest – and these almost always include Japanese language patents.  But it’s also important to understand which search tools are being used by Japanese-language patent searchers, and whether they have access to high-quality full text data.  Recently I was fortunate to catch a demonstration of CKS Web, a Japanese-language interface which can search a database of full text Japanese patent documents in their native language.

According to CKS representatives, although the product offers full text produced by Optical Character Recognition (OCR) from the date range 1986-1993, their collection has been produced relatively recently, which means that it may benefit from advances in OCR technology when compared with older collections.  Another advantage is that patent drawings are also scanned and “clipped,” meaning they appear in the product alongside the patent text; users do not have to download the document image in order to review the patent drawings.

Perhaps one of the biggest highlights of CKS Web was the fact that the search system includes Japanese legal status data and re-assignment data directly from the Japanese Patent Office.  This is an advantage over the multi-country search tools that we frequently discuss on this blog, which rely primarily on the INPADOC legal status file as a source. (The exception is Questel products, which benefit from that company’s relationship with the Patolis Corp.)

Furthermore, CKS Web applies their own proprietary algorithm to generate predictions of the status of each patent document, i.e. whether they are “live” or “dead.”  The product displays this data on the search hit list, making it possible to scan the list for live documents.  With something like this it’s always best to be cautious, as the Japanese Patent Office is really the only entity that can make that determination.  However, the feature may still be of use in certain situations.

Of course, the quality of the search tool is only one of the ingredients necessary for a successful search.  It’s also essential that your native-language searcher is trained in search strategy formulation, a skill that comes with experience.   The provider of this blog, Landon IP, is one company that offers high quality native-language Japanese patent searching with full text coverage.

For a more complete description of CKS Web, including screenshots, please see our Intellogist Community Report on CKS Web.  If you’re familiar with CKS Web, you can register on Intellogist (for free!) to add your own comments about the system.

What are your experiences with native-language Japanese searching?  Let us know in the comments!

PatBase Advertisement

This post was contributed by Intellogist Team member Kristin Whitman. The Intellogist blog is provided for free by Intellogist’s parent company, Landon IP, a major provider of patent search, technical translation, and information services.

Another advantage is that patent drawings are also scanned and “clipped,” meaning they appear in the product alongside the patent text such that users do not have to download the document image in order to review them.

5 Responses

  1. Good initial classification should help even if the shiny bits aren’t in the abstract or title. There’s a reason why the abstract is in front of you, because it’s relevant in some way.

    Here’s patent searching tip #1…consider everything relevant until you find a reason to exclude it,or

    “Look to exclude…don’t look to include”

    • insomniac, my read on the post was that it’s important to go beyond the abstract–as relevant as it might be– because things like classification search (as you point out) and full text search are necessary tools that reveal additional opportunities that would otherwise be missed. If you only consider abstracts (not saying you personally do) then you’re not “consider[ing] everything relevant” from the start.

      Do I understand your point correctly?
      -Chris

  2. Hi Chris

    I think I started to say one thing and ended up saying something else.

    I think I’m saying that whatever you choose to be in your search strategy is there because you or the client think it’s relevant, and it doesn’t matter where or what that might be (abstract, full text, ipc, etc) because if a record (I said abstract before which was the wrong thing to say) resulting from the search is in front of you then it is in some way relevant, and you need to look deeper to find a reason to exclude it.

    So for example if I had a JP specification, I know that the IPC or F-term says it’s relevant even if nothing is apparent from the title or abstract, and if I can’t exclude it on the available information then I need to use a tool as Kristin describes to do the deeper digging.

    I chose classification initially because it’s a little more universal than written language and translations thereof.

    I guess I am saying what you say but from a good search technique viewpoint not from a good tool viewpoint. Great tools don’t make great searchers.

  3. Hi insomniac, you make a great point – “great tools don’t make great searchers” – it’s so quotable!

    I completely agree with you. There is no substitute for using a great searcher. Of course, if a great searcher doesn’t have access to all the data they need, that’s certainly going to hamper their progress. Actually Intellogist was initially created to promote the discussion and evaluation of coverage and features of various search products, because one of the skills of a great searcher is knowing what tools to select for the search.

    Your point is definitely relevant to native-language searches. One of the things we have put a lot of emphasis on at Landon IP is the search quality. Although we have searchers in various languages and various technical backgrounds, we expect consistent quality and training plays a big role in that.

  4. And of course, as we all know, patent searchers must commit to lifelong learning! :

    https://intellogist.wordpress.com/2010/09/09/1403/

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: