Legal

Apply cutting edge A.I. approaches to integrated legal corpora

Connect to patent data sources with one click and analyze them alongside other documents and sources
Communicate with colleagues with secure cortex chat
Maintain lists of important concepts and entities such as people, companies, locations and events

Legal Use Cases

Precedence Search

Often while doing case research, attorneys need to identify precedence for a current case across a wide range of unstructured sources. It is difficult to pull all of those sources into a single analytical platform. It is also exceedingly difficult to identify precedence for a current case if words and phrases don’t match with past case information.

Cortex addresses these challenges by enabling the simple, rapid integration of large sets of documents from a variety of sources without the need for warehousing or re-formatting. Furthermore, Cortex enables the user to apply cutting-edge AI-based text analytics to find words, phrases and documents in sources that share similar meaning and grammar to queried sentences without any keyword or phrase overlap. This allows for far more powerful and effective precedence search and a greater opportunity to find the right needle in the haystack.

Patent Prior Art Examination

Searching for prior art in patent law cases is a challenging problem that requires sifting through large corpora of unstructured content to find critical words, phrases and documents that are relevant to a case. Keyword searching and reading through sometimes thousands of documents is time consuming and prone to error and oversight.

With Cortex, users can apply advanced deep learning analytics to Big Data-scale text corpora that have been integrated into the platform to search with greater efficiency and effectiveness for the critical insights and phrases relevant to their case. Cortex can detect the similarity of sentences even if there is no keyword overlap between them, helping the user identify similar meaning in text. In addition, Cortex’s Neural Concept Recognition AI technology can train on concepts relevant to a case (e.g. people’s names, company names, technical concept names) and scan the entire information corpus to find novel instances of each of those types of concepts.

Concept Centric E-Discovery

Cortex’s Neural Concept Recognition AI technology can train on concepts relevant to a case (e.g. people’s names, company names, technical concept names) and scan the entire rest of the information corpus to find novel instances of those concepts. This allows users to train Cortex on the kinds of things they are interested in finding in a large text corpus and allow Cortex to then go find novel instances (e.g. new people, new companies, new technical terms). For e-discovery this enables users to “look outside the light under the lamp-post” and expand their research to the full universe of concepts relevant to a case.