Using AI for matching patents with technical standards across the Mobile industry.

Woman wearing a smartwatch holding a smartphone

The challenge

Deeper Insights worked with a global organisation who represent leading mobile operator brands. The company wanted to find a better way to map and extract sections of a patent to relevant sections in a standard.

Their current method of extracting patents consisted of expert patent attorneys manually annotating hundreds of documents. This was time-consuming, taking highly skilled workers one hour per document, and causing a bottleneck to scaling the data creation process, or indeed establishing a desired automated end-to-end system.

Deeper Insights have developed data scraping and automation tools using Natural Language Processing (NLP) and Machine Learning (ML) techniques to uniquely identify features for extraction within an unstructured document.

The solution

Using these technologies as well as custom engineering and data science approaches enabled Deeper Insights to create a patent and standards dataset that could be trained to automatically find relevant patents, and sections within patents, that are predicted to be essential to a standard.

Candle chart
Hand holding smartphone

The end result

The solution has enabled the goal of providing an industry-ready service for the exploration of data and analytics for technical standards and related Patents to become highly automated, slashing manual processing and saving many thousands of hours.

Contact us now