Abstract: In various example embodiments, a system and methods are presented for converting query structures for information retrieval from graph-based data structures. The systems and methods receive a natural language query including a set of terms and generate an intermediate semantic relationship of the set of terms of the natural language query. The systems and methods generate a graph query including graph terms corresponding to the set of terms of the natural language query defined by a graph database. The systems and methods search one or more datasets associated with the graph database using the graph query and return a set of results based on the graph query.

Abstract: In various example embodiments, a system and methods are presented for generating node clusters from a plurality of disconnected nodes and generating data access models for interaction with the nodes. The system and methods identify one or more datasets associated with a first set of nodes distributed across a plurality of node clusters, and a set of entities associated within the plurality of node clusters. A node layer is generated based on the one or more datasets and the set of entities. One or more connections are generated between the first set of nodes and a set of coordinating nodes, and between the set of coordinating nodes and a second set of nodes. The systems and methods generate a result set distributed across the plurality of nodes based on connections between the set of coordinating nodes and the first set of nodes and the second set of nodes.

Abstract: Methods and apparatus to facilitate annotation projects to extract structured information from free-form text using NLP techniques. Annotators explore text documents via automated preannotation functions, flexibly formulate annotation schemes and guidelines, annotate text, and adjust annotation labels, schemes and guidelines in real-time as a project evolves. NLP models are readily trained on iterative annotations of sample documents by domain experts in an active learning workflow. Trained models are then employed to automatically annotate a larger body of documents in a project dataset. Experts in a variety of domains can readily develop an annotation project for a specific use-case or business question. In one example, documents relating to the health care domain are effectively annotated and employed to train sophisticated NLP models that provide valuable insights regarding many facets of health care.

Abstract: A graph-based data storage and retrieval system in which multiple subgraphs representing respective datasets in different namespaces are interconnected via a linking or “canonical” layer. Datasets represented by subgraphs in different namespaces may pertain to a particular information domain (e.g., the health care domain), and may include heterogeneous datasets. The canonical layer provides for a substantial reduction of graph complexity required to interconnect corresponding nodes in different subgraphs, which in turn offers advantages as the number of subgraphs (and the number of corresponding nodes in different subgraphs) increases for the particular domain(s) of interest.

Abstract: Methods and apparatus to facilitate annotation projects to extract structured information from free-form text using NLP techniques. Annotators explore text documents via automated preannotation functions, flexibly formulate annotation schemes and guidelines, annotate text, and adjust annotation labels, schemes and guidelines in real-time as a project evolves. NLP models are readily trained on iterative annotations of sample documents by domain experts in an active learning workflow. Trained models are then employed to automatically annotate a larger body of documents in a project dataset. Experts in a variety of domains can readily develop an annotation project for a specific use-case or business question. In one example, documents relating to the health care domain are effectively annotated and employed to train sophisticated NLP models that provide valuable insights regarding many facets of health care.

Abstract: In various example embodiments, a system and methods are presented for converting query structures for information retrieval from graph-based data structures. The systems and methods receive a natural language query including a set of terms and generate an intermediate semantic relationship of the set of terms of the natural language query. The systems and methods generate a graph query including graph terms corresponding to the set of terms of the natural language query defined by a graph database. The systems and methods search one or more datasets associated with the graph database using the graph query and return a set of results based on the graph query.