How Big Data Impacts the World of Patents

Patents and Intellectual Property are gradually gaining significance around the world. This is leading to a bottleneck–large databases and ever growing information. Read on to find how Big Data analytics can show a way out.

To succeed in today’s hostile business environment, innovation is no more a nice to have strategy, but an essential requirement. For the older companies where status quo was acceptable and in fact wasn’t too unprofitable either, things have changed. Smaller companies have pushed the boundaries by providing outstanding products, and adding value at a lower cost. Hence, big enterprises are facing stiff competition. Different strategies have begun to appear in the market. For example, acquiring a brilliant startup with its employees and subsequently focusing on introducing new product lines. Companies such as Yahoo, HP, and Facebook have taken the lead in adopting this strategy.

Another way around the innovation problem is to acquire patents. With examples such as Nokia, Motorola, Twitter, the patent purchases seem rather straightforward. Nokia sold a large chunk of its company to Microsoft, but held on to the crucial patents by signing a licensing deal. They can now earn a revenue using patents licensed to Microsoft. Google bought Motorola and its patents and later sold the company to Lenovo while holding on to the patents. There are ample such examples in the industry. However, there is a bigger game at play behind all this. With millions of patent applications floating around the world, how does one look for a worthy patent. Even if you find one, how do you determine its strength and defensibility (in case of a litigation). Check this page for more information about obtaining patents and avoiding patent infringement read here.

Big Data and Intellectual Property

Transactions of Intellectual Property (IP) is a complex game. Let’s see how Big Data analytics impacts the IP industry. A basic component to be verified before a patent is granted, is novelty. In other words, if a prior-art describing the invention is found, the application stands to be rejected. A prior-art could be in the form of a publication, a blog post, a lecture, a video, or a book. With a humungous amount of information generated—that doubles every 18 months—how does one look for prior art? Sure, a google search will point you to a 100,000 documents, but you’d have to browse through all of these.

One way, some organizations follow, is crowdsourcing the prior art search. The process is rather simple. Details about the patent is published on a website asking IP professionals from around the world to find a prior-art. The winner is offered prize money. This method can and does work to some accuracy, but has its limitations. The emergence of Big Data analytics, on the other hand, has provided a clear way out. The 100,000 documents and hundreds of millions of words can now be intelligently skimmed through using algorithms. In addition, the outcomes through this method get better and precise with each operation.

Let’s talk about the patent issuing authorities. Since Big Data analytics is still not commonly used by most government authorities, prior-art gets overlooked and many bogus patents are granted. This comes out when—in litigation—the opposing parties put all their efforts in looking for a prior-art to invalidate each other’s patents. More often than not, a prior-art is found or there is an out of court settlement.

Hence, a concept called patent wall has gained traction. It is very common for companies to file as well as acquire a number of patents around the technology they are working on. This serves as a defense against litigators and allows the companies to market and sell their products/services without any fear of litigation.

Big Data and Citations Analytics

Another component, where Big Data analytics can be effectively used is Citations analysis. Citations are references provided at the time of filing a patent. For example, if I’ve invented a path breaking new type of a mouse trap, in my patent application, I’d cite other patents issued for mouse traps and prove why my invention is different. Hence, these citations will be linked to my applications as backward citations. Now, imagine if someone comes up with a new way of catching mice and cites my patent in her application. This patent will become a forward citation for my patent. In the patent databases, there are several of these citations (backward as well as forward). Subsequently, this creates a crisscross of citations leading to a complicated web. If you were to study this mesh of forward and backward citations, which is essential in patent analytics, you’d be trapped. But advanced analytics tools can provide you easy-to-understand insights arising from this web of connections.

As the amount of data grows, and more countries see the benefits of a patent system, this information is bound to witness exponential growth. Hence, technologies and initiatives that target the processing of patent data and break it down to insights that humans can understand, will lead the way.