IP Landscape for Machine Learning Patents
What is an IP Landscape?
An intellectual property landscape is a thorough overview of current patents and practices in any technical space. By providing references, documentation, and a holistic view, an IP landscape can show potential for legal validity and inform strategy for design and production of undeveloped products. In short, a patent landscape is a map of everything competing in your area of focus and the current relationships in the field.
The patent landscape can provide a competitive analysis and uncover “white space”, opportunities or unfilled gaps in the current market. Looking at the applicable landscape in the early stages of product development can reveal weaknesses of competitors, advantages you possess, and strongholds or key players that you don’t want to challenge. Detailed examination of the existing scene provides insight into what differentiates your product, establishing focused marketing and transparent communication with investors. This research can also expose what methods have proved valuable and what strategies need adjustment, saving you from known blunders. It is argued that many patents contain information that is not published anywhere else, making an IP landscape the perfect way to obtain new insights in the area of focus.
Pulling a patent landscape starts by looking into a wide variety of relevant data, documents, and information, often found with computer programs or software. Then, real human intelligence is used to narrow down bulk information by sorting it into groups and filtering out irrelevant data. The previously sorted groups are then further examined to find the most important and informative material. The best insight from landscapes is found in the balance of hard data and the story the data tells, meaning a human touch is vital to ensure quality outcomes.
Learn more on how to conduct a patent landscape analysis.
The IP Landscape of Machine Learning From a Patent Perspective
Depending on the industry and focus, the IP landscape can differ widely, making it all the more important to do landscape research specific to your needs. A look at the machine learning patent landscape shows that Microsoft and Alphabet currently hold the most – and highest quality – patents. However, the data also shows that leadership positions can change very quickly in this rapidly developing technology field. In particular Chinese companies are among the most prolific filers in this space, nevertheless their patents fall short in terms of impact and quality.
An analysis with LexisNexis® PatentSight® analytics platform using technology fields defined by ip-search, a commercial patent search service provider of the Swiss Federal Institute of Intellectual Property, provides a detailed look at the machine learning landscape from a patent perspective.
With regards to patent filings the field has seen a tremendous growth over the last two decades. From 2000 the number of applications began to pick up, but the rate of growth accelerated from around 2010 and it is unlikely that this has peaked. Over half of all these patents fall under International Patent Classification (IPC) subclasses G06F (Electrical digital data processing), G06N (Computer systems based on specific computational models) and G06K (Recognition of data; presentation of data; record carriers’ handling record carriers).
Portfolio Size of Machine Learning Patents by Year of First Filing
Unsurprisingly, tech companies are at the forefront of the boom. However, in recent years a large proportion of the first filers have originated from China; among the top 20 machine learning patent filers of 2016, the majority were Chinese companies or institutions.
Top 20 first filers in machine learning in 2016
However, a different picture emerges when examining the top 20 patent owners, proving that isolated data is not as clear cut as it appears. When looking to create/develop the full IP landscape, multiple perspectives must be analyzed against each other. The table below lists the leading companies in machine learning based on Patent Asset Index™ – a measure for the innovative strength of an enterprise, a specific business area, or an entire technology field – as well as total portfolio size. What is immediately apparent is the drop-off in Chinese companies, with all but one – Baidu – disappearing. So, behind the flurry of Chinese filing activity, there appears to be a lack of quality.
Learn more about the Patent Asset Index.
Top 20 Owners in Machine Learning With the Highest Qualitative Strength Measured by Patent Asset Index
However, when looking at each company’s total share of the patent market, most are being outpaced by the rate of development of the technology field. Both Microsoft and Alphabet are exhibiting strong growth in absolute terms, but when looking at relative growth, it is only Alphabet that is increasing its share of the field. Other notable names that are growing their share include IBM and Intel.
Review Alphabet’s patent landscape.
Share of Patent Asset Index of All Machine Learning Patents For Top 10 Owners
In terms of those citing Alphabet’s patents in this field, a significant number of established players – such as Microsoft, Qualcomm, Facebook, Amazon and Samsung – make reference to work done by Google’s parent company. So, while Microsoft currently leads the pack, Alphabet is the main company to keep an eye on.
That said, even Microsoft and Alphabet each take less than a 5% share of the field, while the next eight leading players each take between 1% and 2%. In other words, the machine learning field is still relatively nascent and we are far away from a situation where the top 10 companies are way ahead of anyone else. Because any IP landscape is taken as a snapshot at a specific point in time, all landscapes must be continually monitored for accuracy if they are to be used as a reference or seen as an advisor. Thus, we could yet see a dramatic shift in the positions of the leading players or the rise of new challengers in this highly lucrative and competitive space, especially from Chinese contenders.
As a leading global provider of information and analytics, PatentSight shines a spotlight on the innovations of the past and present and help create data-based projections of what the future might hold. Learn more.
The Patent Asset Index of a patent portfolio is defined as the aggregate strength of all the patents the portfolio contains. See how it works.
Would you like to get more insights into PatentSight Business Intelligence?
Please fill out the following form. We will get in touch with you to arrange an appointment for a webmeeting.
We would love to get to know you and show you some invaluable insights into the IP landscape:
• Let’s have a conversation about your current priorities in IP analytics
• Get an external view on your portfolio
• Learn how others see you and survey your market companions from a new perspective