Noble CoderNamed Entity Recognition (NER) engine for biomedical text
Noble Coder is a generic tool and API that auto codes free text with concepts from controlled terminology. Its algorithm is similar to IndexFinder or ConceptMapper, but unlike these systems, NOBLE Coder uses NoSQL for storage which enables it to code with huge terminologies while keeping its in-memory footprint small.
Noble Coder is part of the Noble Tools Suite, a set of Natural Language Processing (NLP) tools and Application Programming Interfaces (API) written in Java for interfacing with ontologies, auto coding text and extracting information. The Noble Tools suite also includes a generic ontology API for interfacing with Web Ontology Language (OWL) files, OBO and BioPortal ontologies and a number of support utilities and methods useful for NLP (e.g. string normalization, ngram and stemming)
TIES uses this technology to enable quick and easy searches to bring up the results that you want. This technology also uses negation values and specific words to bring up better results in TIES.
How to Use
- When you first download and run Noble Coder, you will be given the option to Input Terminology. This is the dictionary that you want to use to extract specific concepts of interest in unstructured text. You can choose from several different ones found on the Noble Coder site or use your own.NOTE: The first time that you run Noble Coder, you will be prompted with automatically installing NCI Thesaurus for the Input Terminology field.
- You will also be given the option to Input Directory and Output Directory. These options allow you to insert other documents to be processed for extraction.
- Next press the Process Documents button to generate a report of key words and phrases.
- The generated report will appear in a web browser that you can then click on individual files to pull up the main terms located in the document(s). By clicking on the Noble Coder Output [TSV] you can view all of the match terms, codes, concept names, etc.
- In addition to human readable interactive HTML output, Noble Coder generates a tab-delimited text file that can either be loaded into Excel or further processed by a third party software, depending on the use case.
For more information about Noble Coder and the other Noble Coder Tools, please visit: http://noble-tools.dbmi.pitt.edu/
One of the important features of Noble Coder is the ability to build custom terminology databases. These databases allow you to specifically create and establish words and key phrases that make searching through medical documents or other important documents easier.