Moonstone

New to NLP:

If you are unfamiliar with NLP, Moonstone was written to support a wide variety of NLP applications, ranging from applications requiring simple keyword-to-concept mappings ("air hungry-->dyspnea"), to applications requiring deep semantic analysis of narratives, knowledge derived from semantic ontologies (UMLS), rule-based inference (if concept pneumonia is present, flag document for pneumonia), and integration with external NLP tools (eHOST).

Know NLP:

If you are familiar with NLP, Moonstone was written to support a wide variety of NLP applications, ranging from applications requiring simple keyword-to-concept mappings, to applications requiring deep semantic analysis of narratives, knowledge derived from semantic ontologies, rule-based inference, and integration with external NLP tools. The core of Moonstone is a typed first-order knowledge representation language called "MSL" (“Moonstone Semantic Language”). MS output is formulated as MSL expressions which can be queried, used in rule-based inference, used to trigger actions such as sending alerts to clinicians, and so forth.

System development was funded by NIH grant NHLBI 1R01HL116522-01A1

Authors:

Lee Christensen

Wendy Chapman

Associated Institutions:

University of Utah

Intended Users:

NLP researcher or developer

Operating Systems:

Mac

Programming Languages:

Java

Minimum Requirements:

Java 1.7

NLP Task Performed:

Information Retrieval

Information Extraction

Knowledge Base Dev

Download: