Adaptive Information Disclosure (AID)

Partcipating in the VL-e project

Adaptive Information Disclosure (AID) header image 1

NER

(copied from Trac and Javadocs on 06/02/2008)

Javadocs:

NErecognize

public String NErecognize(String input_data,
                          String r_type,
                          String input_type,
                          String output_type)
The NER service takes input provided by a user and outputs either the annotated input (annotated sentences) or all entities found in the input data. To carry out the named entity recognition step, one of the three models is used.
Parameters:
input_data - an absolute path to the text data
r_type - type of NER model (its value is either News or Medline (Genomics))
input_type - type of the data used (lucene if it’s the output from Lucene or text if it is a plain text)
output_type - type of output (annotation if it is annotated input data, NElist for the list of named entities found in the input data or N3 for the named entities in N3 format for the repository)
Returns:
depending on output_type returns either annotated input_data or the list of named entities found in input_data

Copied from trac:

NERecognizerService
2 (Named entity recognition web service)
3  
4 NERecognizerService provides 3 options - either for annotating biomedical text data (EN_GENOMICS.model
5 and EN_GENES_BIOCREATIVE03.model) or for annotating text data based on the newspaper model (EN_NEWS.model).
6  
7  
8 Input: input_data (String)
9 r_type (String)
10  
11  
12 r_type is type of recognition. It must be set to one of the models, “News”, “Genomics”, or “Biocreative”.
13 “News” corresponds to EN_NEWS.model, “Genomics” to EN_GENOMICS.model, and “Biocreative” to
14 EN_GENES_BIOCREATIVE03.model.
15  
16 Output: annotations in XML format (String)
17  
18 The output of NERecognizer is an annotated text with all annotations presented by tags.
19  
20 Realization: Uses lingpipe, SAX Parser

No Comments

0 responses so far ↓

You must log in to post a comment.