| CPC G06F 40/30 (2020.01) [G06F 3/0482 (2013.01); G06F 9/453 (2018.02); G06F 16/24578 (2019.01); G06F 16/287 (2019.01); G06F 40/169 (2020.01); G06F 40/205 (2020.01); G16H 50/70 (2018.01)] | 20 Claims |
|
1. A system comprising:
one or more processors; and
memory storing computer-executable instructions that, when executed, cause the one or more processors to perform operations comprising:
configuring, by a research assistant tool, a research graph to store research results including one or more evidence links associated with a medical domain, wherein the medical domain is associated with a particular subject of knowledge;
receiving, by a query component associated with the research assistant tool, a selection of one or more databases associated with the medical domain;
configuring, by a natural language understanding (NU) engine associated with the research assistant tool, a semantic parser to use a medical ontology to translate natural language text into machine language semantic representations, the medical ontology defining a set of concepts and classifications of the concepts that represent the medical domain;
configuring, by the NLU engine, a set of semantic indicators, a semantic indicator of the set of semantic indicators defining a relational condition for a relationship between concepts to occur, wherein the relational condition is a criterion that is to occur in order for the relationship between concepts to occur;
receiving an input query defining one or more search parameters associated with a research topic, wherein the one or more search parameters include a specific concept and a relation associated with the medical domain, wherein the specific concept is an explicit search term and includes a medical condition, wherein the relation is a semantic link between the specific concept and one or more concepts, wherein the input query is used by the research assistant tool to determine the one or more evidence links;
identifying, by the query component from the selection of the one or more databases, one or more evidence passages that reference the semantic link between the specific concept and the one or more concepts;
determining, by the natural language understanding (NIX) engine using a semantic parser, one or more semantic interpretations for the one or more evidence passages, wherein the semantic parser translates natural language text from the one or more evidence passages into the one or more semantic interpretations with one or more semantic indicators of the set of semantic indicators:
determining, using the medical ontology, one or more ranked concept clusters associated with an aggregation of the one or more concepts based at least in part on a degree of similarity between the one or more concepts referenced in the one or more evidence passages; and
presenting, via a user device, the one or more ranked concept or proposition clusters, wherein individual clusters of the one or more ranked concept or proposition clusters are presented with one or more interactable links to one or more associated portions of the one or more evidence passages.
|