How to Participate in the Open Source Project - PQAI - Patent Quality Artificial Intelligence

For:

The power of open source is the power of people. The people rule.

-Philippe khan, engineer, entrepreneur, and founder of four technology companies

Against:

Open source is an intellectual property destroyer, I can’t imagine something that could be worse than this for the software business and the intellectual property business.

-Jim Allchin, an American computer scientist, philanthropist, and guitarist.

With such contrasting thoughts, imagining open source and intellectual property in the same space is difficult. But not anymore!

An open-source project called PQAI that’s here to democratize patent search is bringing two extremely opposite notions into the same space.

PQAI is here to provide excellent opportunities for researchers and developers to shape the future of patent search AI through open source.

The irony is that there are many more research papers around ai patent search than actual products in the domain of advanced patent tools. The reason being there are not enough opportunities for researchers to participate in live projects.

When starting from scratch, the short duration of 6 months is not enough for researchers at universities to prepare a usable product. Therefore, PQAI is here to give a strong foundation (code base and data sets), necessary documentation, and well-defined challenges (Ph.D. thesis ideas) to participate in this open-source project.

What’s PQAI? | Patent Quality through Artificial Intelligence

PQAI is a not-for-profit initiative focused on creating an open-source AI-based library of software components to accelerate innovation and improve patent quality.

We believe that establishing an open-source forum of IP tools will drive critical changes in the IP ecosystem, like what Linux did in computing and Mozilla did to web browsers.

The PQAI search engine (a free and inventor-friendly prior art search tool) is one example of how such components can be assembled to create new and useful tools.

The empowering of all inventors with advanced IP tools will drive more diversity and inclusion, which will accelerate the pace of innovation. PQAI believes in transparency and user privacy.

Participation Opportunities to Contribute to Open Source Patent Search AI

Around 14 challenges listed in PQAI’s Wiki section can be excellent Ph.D. thesis ideas. Here, we are mentioning three challenges to give you a glimpse:

#1 - Quality measure for the invention query

A search query describing the invention is the starting point of the prior art search. However, each inventor may describe the same invention in different ways. As a result, some queries may be well-formed, while some may be ill-formed.

Some queries could be too short to describe the invention. For example, the query “adjustable mechanical keyboard” does not say enough to be very useful for running a prior-art search.

Some queries may be too broad and open to interpretation. For instance, if a pen is described as a ‘marking device’ in the query.

There is a scope for a module that provides the user some feedback about whether the query is good or not and gives out tips to re-articulate it if required.

#2 - Ease of assessing results

A good chunk of the prior-art searcher’s time goes into sifting through hundreds if not thousands of documents.

Imagine getting an idea about the relevancy of a search result without opening the document, just like a trailer for a movie. Taking a call based on the ‘trailer’ of the patent or research document can reduce the search time significantly.

There is a scope for a module that can extract informative passages from the patent’s specification that are extremely relevant to the search query and shows them upfront.

#3 - Search quality and performance

Ontology is the philosophical study of being and related concepts such as existence, becoming, and reality in metaphysics.

For instance, the surface of a table cannot exist without the table. So, the existence of the surface is dependent on the table ontologically.

Domain-specific ontological knowledge can make prior-art searching more robust to different articulations of the same concept by different users.

However, gathering such information through manual or automatic means, and storing and using it in a search pipeline are open problems.

There is definitely some scope for finding an easier way to gather ontological knowledge and a module that uses this knowledge to make the search more robust.

Parting Thoughts

PQAI is still a baby, and researchers have tremendous scope to contribute to its future.

The challenges belong to different stages of the search, from helping the user create a well-formed search query to finding extremely relevant results across various domains to making it quick and efficient for searchers to go through the results.

The opportunity is worth a quick grab for researchers and developers to participate in a live open-source project!

You can access the PQAI Github directory for further details.