An intelligent search engine that thinks like a patent professional: IPRally offers a brand new way of searching patents - Knowledge Graphs combined with supervised deep learning Artificial Intelligence. Searching, and finding, crucial patent information has always been difficult, and it has traditionally been a job for experts. Comprehensive lists of keywords and synonyms, deep knowledge of Boolean operators and classification systems have been requirements. AI is changing this, and we are changing this. We have already made patent search and review processes more efficient for PTO’s and some of the most innovative companies around the world, and IPRally has allowed them to both save time and increase their accuracy.
So how do we do it? First, we translate the object of your search, be it an existing patent or free text (i.e. draft claims or an invention disclosure), to a graph that maps out the technical features of the text, as well as their relations to each other. Then, this graph is mapped to the 80 million+ graphs that we have already extracted from all other full text patents in our database, and our AI will find the most relevant documents based on technical similarity. The AI is trained using real patent examiner data and is taught to identify similarities like an examiner would.
The result is that you can focus on the core of the technology of interest and let the AI do the tedious feature mapping with a deep level understanding of technology. Patent searching and monitoring has never been easier.