Two tools are better than one - why combining Ambercite with other tools can improve search quality and productivity

Dec 18 2020 A recent blog has discussed the different types of patent search approaches (as shown in the table shown below), but then suggested that a combination of approaches should produce the best outcome.

Method	Boolean (Conventional)	Semantic	Citation searching
Principle	Returns a set of patents that meet a particular query, which often includes a combination of keyword and class code terms.	Searches for patents that have similar keywords or blocks of text	Searches for patents that are similar to one or more starting patents.
Strengths	Well accepted. Can be used for new inventions. Many existing vendors.	Can start with a description of the invention or a representative claim Results are ranked	Excellent ability to return similar results. Results are ranked, and can include relevant patents that have not been previously cited.
Weaknesses	Creating queries can be an artform. Will often return many results that are not relevant, i.e. ‘false positives’. Users can be caught out if relevant patents do not include the expected keywords or class codes. Results may not be ranked in order of relevance.	Can return false positive results, despite having similar results.	Need a starting patent, which may not apply for new inventions. May not pick up relevant patents where citation data is lacking.
Examples	Free: Google Patent, Patent Lens and Espacenet, plus national patent office sites, for example USPTO, IP Australia, and others. Subscription: Patseer, Derwent Innovation, Patbase, Questal, Patsnap.	Innography, IP Rally, Innovation-Q, IP-Screener, Tekmine, Octimine, Incopat.	Ambercite

This has been confirmed by a recent study by Canadian patent search specialists Riahi Patents. Riahi developed a search quality score for 10 different inventions (A to J), and showed that a combination of patent search approaches led to an improvement in search quality compared to conventional searching alone:

When converted into percentage improvement, improvements ranges from 12% to 46%, with an average improvement in search quality of 25%:

These results do not surprise me, after many years of using Ambecite in combination with conventional Boolean patent searching. But how and why is the reason for this?

A case study might help show this.

How can Ambercite searching be combined with Conventional searching?

Lets say you were asked to find prior art for a pulsed blender, similar to the one claimed in US7581688, filed by Whirlpool for a Blender with Crushed Ice Functionality

What prior art would you find for this invention? We already have one answer for this question, as the examiner for this patent reported two examiner citations:

US20060202070, Ice shaver/blender control apparatus and method

US20060203610, Blender control apparatus and method

And what other relevant prior art is there for this invention? We have an answer to this as well, as this patent has been invalidated in the US Federal Court, who found it invalid in the light of US6609821, filed by Sunbeam for a Blender base with food processor capabilities.

But would Ambercite have found this clearly relevant patent?

We always recommend that Ambercite is used to double-check existing search results, and this is what the prior art search double-check would look like, based on the two prior art patents that the examiner cited:

If we run this patent search, this will return 50 potential prior art patents, with the top ranked patent being the Sunbeam patent (click on this image for a fully interactive set of results):

So in this case, an Ambercite search would have easily found this clearly relevant patent, and ranked it that the top of its list.

As an aside, this Sunbeam patent is listed as a citation for the Whirlpool patent, but only as an applicant citation. We have previously reviewed this case, in particular discussing how some people regard examiner citations as superior to applicant citations (on average this may well be the case), but how applicant citations should be not be dismissed out of hand, as they still can be very relevant.

Why does Ambercite improve conventional searching?

Conventional Boolean searching relies on a series of search limiting assumptions about the best keywords and class codes - and these assumptions are necessary, as otherwise the search would never finish.

Ambercite makes no such assumptions, and instead finds patents based on their citation overlap with patents that you have already identified as relevant. By removing the need for such assumptions, this can challenge or at least double-check your existing results. You may find unexpected keywords, class codes or disclosures found in the body of the documents rather than the key text fields.

How does this improve patent searching productivity?

From long experience, a patent search can often find one or more relevant patents in a reasonable period of time, but it can take much longer to find a larger and more complete set.

Or maybe you have been to asked to invalidate a known patent, or already know some relevant patents.

Once you have found one or more relevant patents, you can use Ambercite to quickly expand the data set of relevant patents. These patents can contain unexpected and relevant keywords, or may even be better than the patents you have found (and you can use these patents to expand your search query in an iterative searching strategy).

After you think you have exhausted a conventional search, you can then use Ambercite to double-check your final set of results - you may find new and relevant patents, or you may decided that your set is comprehensive. Either way, you are better informed and in a better position.

All these benefits can help improve patent searching productivity.

Do you want to test these features and benefits for yourself?

Ambercite offers free trials, but to get the most of this, please contact us for a demonstration. You can try either option via the links below: