Discussion of all things patent mapping and analytics.
The US based National Security Agency (NSA) has been in the news a lot recently due to its activities, which include the collection and analysis of telephone and internet records. While the exact nature of this analysis is both confidential and controversial, it is thought to include analysis of websites visited, and links between people as expressed via email and message traffic, and calls made. This type of analysis is often referred to as 'metadata' analysis, where metadata can be defined as 'data about data'.
But is not only the NSA that understands the value of metadata. As discussed by Maria Konnikova's excellent article on MIT's human dynamics scientist Dr Alex Pentland, the likes of Google and Apple are also interested in metadata. Pentland's work began with counting Canadian beavers from outer space back in 1973. Since beavers were small, and satelites crude back in 1973, Pentland started counting beaver ponds instead. This is a great example of the use of indirect measures to try to find data that would otherwise be impractical to gather. Since then Pentland has been working on wearble technology like Google glasses, as well as on metadata in general:
The thing is, I can read most of your life from your metadata," Pentland says. "And what’s worse, I can read your metadata from the people you interact with. I don’t have to see you at all.
Some might argue that this is scary stuff. But what has this to do with patents?
Metadata has long been used to help search for patents. Patent classification codes are a form of metadata, as are owner names. But patent classification codes can be very inaccurate, and company names can be limiting if you don't know who is filing that patents you are looking for. Meanwhile other patent search techniques such as keyword searches can miss relevant patents if a different and unexpected terminology is used.
There is another type of metadata analysis that can be applied to patents. Similar to the words of Pentland above, using the very valuable metadata found in patent citation links (which are interactions between patents) can determine similarity between patents, even when this similarity has been missed by others in the field. For example, consider how:
Some of the concepts used in these analysis to those believed to be similar to those being applied by the NSA to analyse phone and internet traffic (but to avoid doubt, Ambercite will only apply its algorithms to publically available patent and patent citation data!)
There are many other examples we could name about how Ambercite and its search system AmberScope has been used to find patents that made a major difference to our clients, as our clients are happy to tell us:
"My role is more directly analyzing patents including claims, infringement and prior art. AmberScope appears exceptional for identifying prior art from anything else I've seen." - Paul Morinville, CEO at OrgStructure, US
Interested in applying metadata analysis to your patent searching?
Ambercite provides a number of options for this:
1) AmberScope -
Search the patent network using an inutitive interactive web app (click on the image for more information)
2) Automated patent searching
Our algorithms analyse patent citation data to determine the most similar listed and un-cited prior art.
3) Network Patent Analysis (NPA)
A whole series of advanced algorithms rank and group all of the patents in a targeted area of technology
Click on any of the above images, or contact us directly to find out how to apply advanced metadata analysis to assist your company.