Measuring the Accuracy of ML and AI for Classifying Patents

Measuring the Accuracy of AI and ML for Classifying Patents

Watch recording

We hosted a webinar in collaboration with Patinformatics LLC around how we can test ML (machine learning) for its accuracy in classifying patents.

The webinar stemmed from an article published by Cipher in World Patent Information, Construction and Evaluation of Gold Standards.

It aims to provide an objective and reliable way for both developers and users of patent classification software to better understand and test the accuracy of automated patent technology mapping.

We were joined by the Managing Director at Patinformatics LLC, Tony Trippe, and our very own CTO, Steve Harris, who spoke about the Gold Standard for measuring the accuracy of machine learning and AI.

Topics covered

  • How a Gold Standard helps test the accuracy of machine learning
  • The approach to building a Gold Standard that can be trusted
  • Testing Cipher against the Gold Standard



Please fill out the form with your details, and we will get in touch to schedule a demo.

Request a Demo of Cipher Classifiers

We are excited to show you an alternative to manually tagging patents and building long boolean queries to get your view of the technology landscape. With the automated classifiers, you can:

  • Pull through all the latest patents in the technology area
  • Analyze the results to understand new technology trends
  • Benchmark your portfolio against your competitors 
  • Optimize your portfolio for monetisation or risk management

Submit the form, and we’ll be in touch.

Was this post helpful?