JCoRe OpenNLP Tokenizer · UIMA wrapper for OpenNLP Tokenizer. The difference to the official OpenNLP UIMA wrapper is basically the type system. In this project the JCoRe type system is employed and is thus compatible to other JCoRe components.
JCoRe Open NLP Chunker · UIMA wrapper for the OpenNLP Chunker. The difference to the official OpenNLP UIMA wrapper is basically the type system. In this project the JCoRe type system is employed and is thus compatible to other JCoRe components.
JCoRe Mantra XML Types · Additional types for the Mantra Challenge (see iexml-{reader,cosnumer}).
JCoRe Token Annotator · UIMA Wrapper for the JULIE Token Boundary Detector.
JCoRe Sentence Annotator · UIMA Wrapper for the JULIE Sentence Boundary Detector.
JULIE Lab Neo4j Concept Management Plugin · This project includes multiple plugins for the Neo4j server to support to import, management and query of concepts.
JULIE Entity Evaluator · A small library aiming at a standardized way to do entity recognition evaluations.
JULIE Lab Embeddings Service Provider Interfaces · This project provides common interfaces used by other embedding classes.
JCoRe XMI Database Writer · Writes CAS data in XMI format into a relational (PostgreSQL) database. It is possible to write the whole XMI for each document into the database. However, it is also possible to define a list of annotations that should be written into separate tables. The JCoRe XMI Database Reader can then be used to assemble XMI data from a given set of annotations.
JCoRe XMI Database Reader · Reads CAS XMI data from a relational database (Postgres). Thus, the stored CASes may then be processed further.
JCoRe Linnaeus Species Tagger · Uses the Linnaeus software (http://linnaeus.sourceforge.net/) to detect species mentions and map them to NCBI taxonomy IDs. This project contains the UIMA wrapper around the original Linnaeus software but does not include any dictionaries. Since Linnaeus works with different dictionaries (e.g. concrete species name mention recognition, species proxy/clue recognition, ...) each of the different resource has been stored in a project of its own in jcore-projects. The default mention dictionary is stored in the jcore-linneaus-species-dict project, for example.
JCoRe MST Parser AE · Slightly modified version of the MST parser to fit into the UIMA framework. This dependency parser determines all syntactic dependencies between words within a sentence. Certain level of adaptation of the original source code was necessary to neatly integrate the parser in UIMA.
JCoRe Lingpipe Gazetteer AE · Basically used as NE tagger based on Lingpipe's dictionary-lookup tagger.
JCoRe BioNLP 09 Event Reader · This reader generates CAS' from data that adheres to the BioNLP-ST (2009) format.
JCoRe BioNLP 09 Event Consumer · This consumer is used to generate BioNLP-ST (2009) compliant data from a CAS.
BioSem Event Extractor · The event extraction engine used by the Amsterdam team in the BioNLP Shared Task 2011 challenge. The code was adapted to also work with all data of the ST2013 challenge (there are 4 more trigger/event types in these data in comparison with the ST2011), evaluation results stayed the same.
JCoRe Token Annotator, Biomedical English · UIMA Wrapper for the JCoRe Token Boundary Detector with a model trained on a special biomedical corpus which consists of data from (manually annotated) material which we took from MedLine abstracts and a modified version of PennBioIE's underlying tokenization.
JCoRe Sentence Annotator, Biomedical English · UIMA Wrapper for the JCoRe Sentence Boundary Detector (jcore-jsbd-ae) with a model trained on data from both the GENIA and PennBioIE corpus as well as additional material from MedLine abstracts.
JCoRe PubMed Reader · Leveraging the JCoRe XML Reader, this project employs a PubMed-enabled mapping file to map PubMed XML documents to CAS instances. The input is a directory of single XML files, each containing exactly one XML PubmedArticle. The XML format is the one used by NCBI when displaying results in PubMed as XML.
JCoRe OpenNLP POS Tagger, Biomedical English · This project employs the OpenNLP wrapper (jcore-opennlp-postag-ae) with a model trained on the PennBioIE corpus.
