Amberscope vs Ambercite Ai - which is better?

Amberscope was the first patent searching product launched by Ambercite, and was well lauded for its unique visualization of the patent landscape, still unmatched by competitors. Insights and feedback from Amberscope then led to Ambercite Cluster Searching, which was more recently replaced by Ambercite Ai, an even more advanced patent searching platform.

Given the progression from Amberscope to Ambercite Ai, does this mean that Amberscope is now obsolete?

Not at all! Both Amberscope and Ambercite Ai have their place, as will be discussed by the comparison below.

To run this comparison, we will look for similar patents to a recently published PCT publication, namely WO2017003624, filed by Intel for an Integration of force transducer into tablet to measure weight – or a mobile device that can weigh stuff.  Claim 1 reads:

1. An apparatus comprising:

logic, the logic at least partially comprising hardware logic, to receive a load detection signal from one or more load sensors coupled to a mobile computing device,

wherein the one or more load sensors are to be integrated into a chassis of the mobile computing device.

 

Which I can imagine would have enormous appeal to the type of younger and entrepreneurial folk that can be found around many city centres.

This patent was published in January 2017, along with five patent citations - and provides an ideal case to explores the differences between the two products.

 

So what additional information can Amberscope and Ambercite Ai provide?

We will start with the earlier product Amberscope. If we enter the number WO2017003624 into this system, we end up with the following screen

 

Note that there is the ‘seed patent’ being WO2017003624, five backward citation patents present as dots that point towards the seed patent (arrows always point from older to newer patents in Amberscope), and a series of ‘ghost like’ patents.

By hovering over any of these dots, further information is displayed, such as shown below:

 

This is one of the listed backward citations – and appears to be relevant.

The ghost patents, btw, are citations of citations. Amberscope does not show them all, but instead selects what may be the most similar patents to the seed patents, for example:

 

 

Which is very similar to the seed patent – yet was not listed as a citation. So this starts to show some of the value of the Amberscope approach.

But this only part of what we can do. For example, consider this patent here:

 

While a direct citation, this appears to be a slightly different approach. Maybe there are more patents like this?

Note that this details box as shown includes the rectangle “13 more’. What this means is that if you were to press this area, Amberscope would be refocused on this patent and 13 additional patents would be shown – with the resulting network looking like this:

 

This network looks a little different to the WO2017 patent, with more backward citations. However few of these patents are connected to each other – this lack of cross-citations can suggest an technical area that is still developing.

Note that this is only part of what Amberscope does – further details can be found here.

 

Ambercite Ai

Ambercite Ai applies many of the same principles to the same data, but in a table form. This is shown below (click on the image to see it in better resolution):

 

This time the most similar patents are listed in table form, in ranked order according to our similarity algorithm. Only the highest ranking five patents out the 42 found are shown in this snapshot.  

There are a lot of columns shown and further information about each column is found here, but one of the key fields is the second to last column in the bottom table, headed ‘Known’. This notes whether the patent (family) found is a known citation   - or not.

In this case, the top two ranked patents are not known citations. But are they relevant?

  • US9091585, which had its first family publication* in August 2014, discloses a smartphone that can be used to weigh stuff, including with a load sensor
  • US9366588, first published on 18 June 2015 or just 9 days before the priority date of the WO2017 patent, discloses a mobile platform with a load sensor (and also an area sensor).

An further down in 5th position is the unknown ‘citation’ US20120181091, for a System and Method for weighing a device on a mobile phone.

 

* To assist with finding prior art, Ambercite when in validity searching mode displays that first date that any member of this patent family was published. In this case the earliest published family member was US20140224551.

 

How do Amberscope and Ambercite Ai compare?

Both approaches have their strong supporters. Users of Amberscope like the visual interface and how this can quickly suggest the number of similar patents. Sometimes the patents form into groups of distinctive clusters, such as shown below, and these can suggest a common subject matter.

 

Users of Ambercite Ai like that the patents found are ranked in order of similarity – and also that you can run a search on more than one seed patent:

 

Up to 200 starting patents in fact, which can be really helpful if you have ended up with short list of relevant patents in a search that you are running, and want to know what else you can find.

You can even exclude other known patent families from the results if you wish:

There is also the ability to preference which family members are shown in the results – for example EP as shown below:

EP preference.gif

 

Maybe you want a bit of both? For this reason you can easily the Amberscope network for any of the patents listed in Ambercite Ai, by selecting the magnifying symbol, as shown below:

Amberscope link.gif

 

Which is better?

I think that this is a bit of a false dichotomy – both system have their benefits, which is why we offer both to our clients. I personally use the Ambercite Ai the most, but also use Amberscope for gaining a visual understanding of the network and for demonstrating the principles of Ambercite.

So the answer is – whatever system you prefer.