Will IP Intelligence Revolutionize Patent Search?
Published on 28 Sep, 2015
About 615,243 patent applications were filed in the year 2014.
That works out to about 1700 applications per day; of which almost 900 made it through.
That’s 900 odd new patents granted, every day, for a whole year.
Those figures are only expected to grow, perhaps exponentially, in the coming decades.
Well, those 900 new patents are added to an already extensive list of over 6 million patents granted since 1963.
That’s a veritable ocean of innovation and intellectual property.
It grows deeper and deeper as mankind surpasses the limits of our knowledge and innovation, constantly pushing the boundaries of our aspirational accomplishments.
While that’s all well and good, wrap your head around a simple fact — as an IP researcher, you’d need to navigate that immense ocean. Traverse its vastness for pearls of intellectual property.
Well, you’re not the only one.
As the global web of protective and prohibitive patents grows thicker by the day, there’s a pressing need for innovative and intuitive approaches to patent research. We’re witnessing a growing need for an "IP Intelligence" of sorts, something that can assist a patent researcher in finding relevant information with ease.
Consider some facts:
- The United States Patent Office (USPTO) generates a whopping 12 GB of compressed patent data every week.
- China is ahead of the US, with its annual volume of invention applications likely to grow well over 900,000 by 2018.
This burgeoning number of applications has led to new and innovative solutions that help analyze patents. Some new companies such as Innography, Ambercite, and Aistemos, have created proprietary IP analytics tools that can help researchers analyze huge volumes of patent data, extracting insights at the click of a button.
However, before extracting any intuitive insights, one of the foremost requirements of any patent-related search is arranging patents in a specified order that suits your unique requirements.
Let’s consider this necessity for arrangement.
While conducting a patent search, you’ll certainly have some criteria in mind. Let’s say you need to look at patents filed before date X, examine patents filed by company Y, check patents awarded to inventor Z, and so on.
You will need to arrange your set of patents in a specific order, or filter them by some specific criteria. Almost all bibliographic data associated with a patent — filing date, assignee, inventor, etc. — can be used to sort patents. Most patent databases such as Google Patents, Thomson Innovation, and Questel Orbit, provide such filters and sorting options. However, as a researcher, there may be times when it seems you’ve run out of options, and you desire other more advanced ways to sort your data.
While performing a patent search, you’ll have three generic objectives:
- Locate exact prior art (to invalidate a patent)
- Locate close and nearby prior art (to validate a new invention)
- Locate certain features in patent claims (to analyze infringement)
Each of these search objectives demands a different set of guidelines and filters to be applied on patent data.
For example, in the third case, you’d be more interested in locating information in the claims of the patent. Similarly, for the first case, you would look for patents filed before a certain date that explicitly mention certain features.
With the amount of available patent data constantly on the rise, applying multiple or varying filters as per specific requirements can be rather cumbersome for a researcher. There is a dire need for intelligence in this aspect.
You’d probably envisage a solution wherein we sort patents by a straightforward filter named "Objective", drawing from the aforementioned three objectives.
An ideal solution would go like this:
- Execute a patent search query and get a result set
- Apply an objective filter
- Sort the patents as per their relevancy with the input query
Such a system could conduct multiple analytics projects at the same time using the same search query.
With the kind of work being done in IP analytics, such intuitive features may soon be commonplace in existing patent databases, heralding a new age of ease and efficiency in intellectual property research and analytics.