Machine Learning Basics – Patent Claim Charts – Case Study

Machine Learning

Machine Learning or ML is a concept that enables a machine or a software tool to learn on its own without specific programming.

For example, in banking sector, ML can be used to predict unpaid or bad loans before a default actually occurs based on the previous payment patterns.


Other uses of ML include image recognition, like suggesting a name of a friend to tag in a particular image, or speech recognition, like conducting searches and performing other tasks after speech to text processing.

Similarly, diseases can be predicted using clinical parameters and medical history of the patient along with the medical history of the family members of the patient.

Machine Learning Models

Generally, the machine learning algorithm is trained using training data to create learning models. Subsequently, when new data is introduced to the algorithm, it makes a prediction. The machine learning algorithm can be categorized into 3 types, supervised learning, unsupervised learning, and reinforced learning.

In case of supervised learning, the input data acts as a reference or a foundation. For example, if data relates to weather, the algorithm can be trained to learn about specific parameters like temperature and humidity patterns.

In case of unsupervised learning, the algorithm determines the patterns on its own. For example, if data covers weather and population information, the algorithm can create separate data clusters for weather and population data, so that when new data is introduced, it can be automatically fed to the correct cluster.

In case of reinforced learning, the tools is programmed to determine the best possible solution on its own for achieving a particular goal, mostly by interacting with an environment and to learn and evolve using such interactions.

A good example of reinforced learning is elevator service. Like in our office, there are 6 elevators to transport people across 25 floors. So, to provide a timely service, the elevator controller manages the elevators on its own depending upon factors like number of calls and direction of each call.

While training the machine using different methods, valuable intellectual property is generated at each step, which can be protected by way of patents.

For example, patents can be filed for new tools to prevent cyber attacks or for speech recognition tools for better processing of natural languages across different geographies.

Such patents can then be licensed to other industries for broader applications.

Patent Claim Charts

With ever-growing number of patents in the sector covering computer related inventions, software patents, mobile app patents, IoT (Internet of Things) and Artificial Intelligence (AI) Patents, claim charts can be prepared to determine patent infringements or for exploring patent licensing opportunities.

In use, patent claim charts can illustrate an overlap between a product (or a service) and the claims of a patent. Essentially, the patent claim charts dissect the claims of a patent by its specific elements. Other names of claim charts that are used by patent attorneys are, patent claims tables, or patent infringement contentions, or patent invalidity contentions, or patent preliminary infringement contentions (PICs).

In addition to drafting a strong patent application, protecting the company’s intellectual property is a crucial business practice. Strong patent and IP protection ensures that the unique elements of company’s products and processes are protected and adds a high degree of credibility to the company in the industry.

Apart from adding to patent licensing potential, patent claim charts illustrate the extent of patent infringement in a simple and visually understandable manner. Claim charts are also known as ways to illustrate patent mapping between the products and the patent claims, which are used to signify “Evidence of Use” for licensing strategies and patent litigation research.

Product to patent claim mapping is a detailed comparison of granted independent set of patent claims with the product, apparatus, machine, system or method / process

Best 6 Steps for Performing Product to Patent Claim Mapping

  1. Identify the infringing products.

  2. Prepare a claim mapping table describing every feature of the independent patent claim in separate cell.

  3. Interpret the scope and understand every feature of the independent patent claim.

  4. Note down the interpretation of every feature of the independent patent claim.

  5. Map (Compare) the identified infringed product with respect to every feature of the independent patent claim.

  6. Highlight the features identified in the product brochure for easy reference.

What is product to patent claim mapping in patent study?

Product to patent claim chart mapping is a detailed comparison of granted independent set of patent claims with the product, apparatus, machine, system or method / process that may be infringing the granted patent claims. The patent claim chart includes comparison of identified products infringing the granted patent. The steps includes analysing a set of granted patent claims with product features. Our computer software attorneys who are trained by US patent attorneys provide product to patent claim mapping services for ongoing US Patent court litigation. Our CS lawyers have been preparing Evidence of Use patent charts for over 10 years.

Product to patent claim mapping services for independent granted USA patent claims

Sample US claim chart for product to patent claim mapping

Patent Claim 16


Feature Present (Y/N)

Product 1 (iDAvatars)

16. An article of manufacture including a non-transitory computer-readable storage medium having stored hereon computer-executable instructions, execution of which, by a computing device, causes the computing device to perform operations for performing speech recognition, the operations comprising:

Any computing device performing one or more operations for performing speech recognition


Intelligent Virtual Assistant (IVA) [Page 1, para number 4]

initiating a finite state machine; User interacting with the computer-implemented voice user interface with personality. Y iDAvatars combines artificial intelligence with cutting-edge natural language processing technologies and state of the art animation to produce avatars that listen, speak and respond through talk, text and touch [Page 1, para number 2] voice response or text-only guidance [Page 1, para number 4]
receiving a voice input; Input data is user speech. Y Conversational avatars that promote patient engagement [Page 1, para number 1]

state of the art animation to produce avatars that listen [Page 1, para number 2]

interpreting the received voice input, comprising: interpret the spoken command of the user Y Natural Language Processing expertise [Page 1, para number 4] Speak and respond through talk [Page 1, para number 2]
transitioning to a domain state functionality of the finite state machine, determining which domain of functionality to proceed to next based upon an user dialog Y Natural Language Processing expertise, Cognitive computing capabilities [Page 1, para number 4]
selecting a generic prompt corresponding to the domain state functionality, and Analysing user dialog and selecting pre-stored response Y Natural Language Processing expertise, Cognitive computing capabilities [Page 1, para number 4]
selecting a specific prompt corresponding to the generic prompt, selecting a specific prompt that corresponds to the generic prompt, the voice user interface with personality considers both prompt history 930 (i.e., what the virtual assistant has said to the subscriber) and recognition history 916 (what the user has said to the virtual assistant). Y Natural Language Processing expertise,Cognitive computing capabilities [Page 1, para number 4]
wherein the specific prompt comprises
a variant of the generic prompt and also corresponds to the domain state functionality; Each prompt includes both a specific name (e.g., a specific prompt) and a generic name (e.g., a specific prompt corresponds to a generic prompt, and several different specific prompts can correspond to the generic prompt). Voice user interface with personality 1002 determines which domain of functionality to proceed to next based upon a dialog (e.g., dialog 1008) with a subscriber. N
and transmitting the specific prompt in a response. Computer-implemented voice user interface with personality can output to respond to various types of questions or responses in various situations during interactions with the user Y iDAvatars combines artificial intelligence with cutting-edge natural language processing technologies and state of the art animation to produce avatars that listen, speak and respond through talk [Page 1, para number 2]

We provide comprehensive Patent and Trademark legal services via our global network to create valuable patent portfolios and resolve complex patent disputes by providing patent litigation support services.

Our team of advanced patent attorneys assists clients with patent searches, drafting patent applications, and patent (intellectual property) agreements, including licensing and non-disclosure agreements.

Advocate Rahul Dev is a Patent Attorney & International Business Lawyer practicing Technology, Intellectual Property & Corporate Laws. He is reachable at rd (at) patentbusinesslawyer (dot) com & @rdpatentlawyer on Twitter.

Quoted in and contributed to 50+ national & international publications (Bloomberg, FirstPost, SwissInfo, Outlook Money, Yahoo News, Times of India, Economic Times, Business Standard, Quartz, Global Legal Post, International Bar Association, LawAsia, BioSpectrum Asia, Digital News Asia, e27, Leaders Speak, Entrepreneur India, VCCircle, AutoTech).

Regularly invited to speak at international & national platforms (conferences, TV channels, seminars, corporate trainings, government workshops) on technology, patents, business strategy, legal developments, leadership & management.

Working closely with patent attorneys along with international law firms with significant experience with lawyers in Asia Pacific providing services to clients in US and Europe. Flagship services include international patent and trademark filingspatent services in India and global patent consulting services.

Global Blockchain Lawyers ( is a digital platform to discuss legal issues, latest technology and legal developments, and applicable laws in the dynamic field of Digital Currency, Blockchain, Bitcoin, Cryptocurrency and raising capital through the sale of tokens or coins (ICO or Initial Coin Offerings).

Blockchain ecosystem in India is evolving at a rapid pace and a proactive legal approach is required by blockchain lawyers in India to understand the complex nature of applicable laws and regulations.


Level18,One Horizon Center,
Golf Course Road
DLF Phase 5,Sector 43

The Bar Council of India restricts any form of advertisements. This blog contains general information for the convenience of readers and does not purport to dispense legal advice and is not intended to solicit or advertise in any manner.
Gurgaon,Haryana 122002

No Attorney-Client Relationship
The use of our blog, and the sending or receipt of information via this platform does not create an attorney-client relationship between you and us.