How AI Can Make Patent Searching Easy: Artificial Intelligence (AI) is a broad concept that has managed to enter our world subtly and benevolently. Although all technological inventions bring a massive change to human life, AI goes beyond just transforming lives. It stands to understand and replicate the whole fabric of thought and existence. In simple terms, AI is nothing but a computer program seeking to learn. Just like humans observe and absorb their surroundings, AI observes and learns through the data it is fed. Read on to know how AI can be used for patent searching.

AI is Everywhere

Have you ever noticed how your YouTube feed always has interesting videos to offer, or how Alexa and Google Assistant can command your house as properly as a butler? There are many examples of AI's presence in our day-to-day life, from weather forecasting to recommendations on your favourite online retail store. AI is essentially a human brain in the making. It is an expert system that will reduce human effort.

Patent Searching

Patent search involves searching existing patents as well as publicly available documents to establish the novelty of an invention, which can be anything – technical concept, machine, algorithms, structures, compositions, designs, etc. Further, this search also helps to determine if any granted patent was already disclosed earlier and thus, challenge the validity of the granted claims.

AI and Patent Searching

There is a steep rise in the number of IP assets being filed globally. According to the WIPO 2019 report there was a 5.2% growth in patent filings between 2018 and 2019 whereas the Utility model application has grown by 21.8% between 2018 and 2019.  This upward trend in filings has continued for at last two decades. Therefore, IP resources have grown year on year and finding relevant information in this vast amount of data is becoming more difficult and time consuming. 

With increasing volume of patent data, finding relevant information and analysing prior arts is time-consuming when done manually, especially if you want to achieve quality and accuracy. AI can enhance patent searching and help you save time, cost, and effort. While performing patent searches, AI provides high accuracy and incomparable quality in a quick turnaround scenario.

Cohesive approach through effective reuse of knowledge

Deep Learning (DL) and Neural Networks (NN) not only automate the process of searching in huge volumes of information but also learn (store and use) from previously analysed data to improve the accuracy of overall searches. Today's research is focused on using deep learning and neural networks for classifying or categorizing the patents and finding similar patents. Natural language processing (NLP) is also being used to suggest contextually relevant keywords and their synonyms. This leads to improved concordance between the available knowledge and documents that users want to search for. AI can also help in deriving Insight into the strengths and weaknesses of a technology segment in certain geographies by cross referencing with the IP data and delivering an instant overview of the domain.

What magic happens in AI based Patent searching tools – there are multiple tools available for the searchers and each tool has features making them unique to use. Some tools not only provide a list of results based on AI similarity algorithms, but rank/score them based on the level of contextual similarity with the target invention you are looking for. Some allow you to do a more meaningful citations analysis and a few of them allow you to search for similar patents in context of more than one target patents when the analysis requires the scattered concept (depicted in two already know patents) to be identified in a single prior art.

Also, in case you believe that any one of the two target patents (whose similar prior art you are looking for) is more important, you can adjust their respective weights and the query results will be affected by your selective weighing as depicted below (from the Ambercite interface):

Also, AI tools are being designed to increase the efficiency of analysis for the searcher e.g., Ambercite remembers the patents you have already reviewed – so you do not need to review them again.

A follow-up query for a search produced 50 results as requested, but only 6 of these were new. By selecting the New Patents as shown below, you can save your valuable time:

Also, there are AI- based search tools which tend to find similar prior art by targeting required concept rather than only the keywords. They use state of the art data science to understand documents and find the connective threads within.

AI tools have a huge appetite for learning just like an infant and is just waiting for us to train them. Amplified AI automatically learns with you as you analyse the results identified by the AI engine. Most of the AI tools have adaptive intelligence to grow their conscience and familiarity about a technology as we perform more and more searches.

Today, IP professionals have access to various patent databases and tools which provide accurate patent search results. Other Patent databases such as Ambercite, Amplified AI, Dorothy AI, Resolute AI, etc. operate on AI to provide analysis and high-performance searching. With improved algorithms, IP professionals are now equipped with better decision-making tools and can effectively reach conclusions through their own analysis and interpretation of the AI recommended results.

Conclusion

The advent of the digital era has led to rapid transformation of numerous industries and sectors. Artificial intelligence (AI) is one digital technology that has heavily impacted various industries. With digitalization and accumulation of patent big data and AI becoming a common processing term due to improved capability of computer to perform tasks such as visual perception, text processing, decision-making, contextual matchmaking, NLP, language translation, etc., AI has progressed rapidly over the last few years in IP industry. Though it still has a long way to match the multi-dimensions of human intelligence, it has still allowed IP professionals to shift their focus to more strategic tasks and make quick use of all the data around them in more structured and intelligent manner.

Author:

Sumit Prasad, Group Manager, ICT Searching, Sagacious IP.