ICmasters AI+ accelerates patent mining

A patent gives its owner the legal right to exclude others from making, using, selling, and importing an invention for a number of years. However, all that glitters is not gold. In the semiconductor industry, it is estimated that only about 5% of patents can provide economic value.

Patent mining is a process of reviewing and analyzing patents to maximize the value of a patent portfolio. To mine patents, a patent engineer must identify the additional innovations from the present technologies in the disclosed patents and interpret the coverage of the patent claims. The patent team is required to draw from their insights and experiences in semiconductor devices and the current state of the industry, as well as a “patent sense” developed from experience in evidence-based patent review and EoU (evidence of use) documentation cases.

However, patent mining is a time-consuming and costly process. Reviewing thousands of patents manually while frequently cross-checking between large patent and device analysis databases is a difficult task for anyone for extended periods of time. In large projects, errors inevitably creep in due to the cognitively demanding nature of the job. High-quality work is difficult to sustain, and high-potential patents are missed. For example, top semiconductor companies, such as Samsung, Intel, SK Hynix, TSMC and Micron have approximate 91,000, 33,000, 11,000, 17,000 and 22,000 valid US patents respectively. Just taking 15 minutes for each patent, mining 17,000 TSMC patents would require an experienced patent engineer to work continually for over 600 days to complete the effort, or 10 patent engineers 12 weeks each.

The ICmasters team is a group of experienced patent mining and analysis experts, as recognized by the world’s top semiconductor companies, intellectual property law firms, and patent holding companies over the twenty years they have worked with us. To strengthen our patent analysis efficiency, ICmasters has created an in-house AI+ platform that integrates our proprietary patent analysis skills, optimized patent mining workflow, and state-of-the-art AI technologies. The AI component of the AI+ platform integrates our engineers’ insights and experience in semiconductor reverse engineering and patent analysis. As a result, the time-consuming process of patent mining and analysis are sped up immensely.

Using the AI+ platform enables our team to accelerate the patent analysis process by a factor of ten times or more, depending on the project. Still taking the 17,000 TSMC patents as an example, it would require two of our patent engineers about one month to finish it.

The ICmasters AI+ platform is not only for patent mining but patent portfolio analysis in all aspects. Taking OLED technology as an example, the ICmasters AI+ platform can pick up a list of Samsung patents that provide the best matches to an LG OLED device under analysis. Similarly, the same proces can be used to find popular Samsung OLED displays on the market that match selected LG patents. It is a much more efficient way for patent engineers to provide EoU documenting that given devices infringe on disclosed patents.

The ICmasters AI+ platform is used to augment, not replace, experienced engineers. It can be thought of as another team of patent engineers under the guidance of senior patent engineers. This and related techniques can be used in a number of associated use cases in patent mining and patent categorization throughout the semiconductor industry.


Bing Liu holds a Bachelor’s degree in Radio Communications from the University of Science and Technology of China. In early career he occupied a series of integrated circuit design roles with a focus on RF, analog and mixed-signal ICs. He later moved his family to Canada and became one of the industry’s most sought-after circuit analysts. In mid-2017, Mr. Liu founded ICmasters to bring best-in-class reverse engineering analysis to the semiconductor industry.

Yuan Liu is a software engineer with over five years of industry experience, including cloud computing, data analytics, and machine learning. With previous experience at the federal government of Canada as well as a freelance software developer, he is constantly on the look-out for new challenges. He joined ICMasters in 2017 to apply his expertise to the most pressing issues of the day in the semiconductor industry.

Source: https://icmasters.com

Related presentations

Liked this post?

Share it on your social networks

Photonics Forums Yole CIOE 2020 1140x370