edu.illinois.cs.cogcomp.annotation

Class XmlTextAnnotationMaker

java.lang.Object

edu.illinois.cs.cogcomp.annotation.XmlTextAnnotationMaker

public class XmlTextAnnotationMaker
extends Object

Instantiates a XmlTextAnnotation object from xml text. The xml is parsed into body text (which is further cleaned up as needed), and this cleaned text is used to create a TextAnnotation. Additional information is extracted from the xml source. The mapping between the xml source and the cleaned text (i.e. mapping between character offsets) is also derived. The goal is to provide text that can be processed easily with an NLP pipeline without a lot of hacks to work around ill-formatted text. The annotations so produced can then be mapped to the offsets in the original xml text, and combined with supplementary information extracted from the xml markup.

Author:

mssammon

Constructor Summary

Constructors

Constructor and Description
XmlTextAnnotationMaker(TextAnnotationBuilder taBuilder, XmlDocumentProcessor xmlProcessor) Specifies the behavior of the XmlTextAnnotationMaker: tokenization (via the TextAnnotationBuilder), which xml tags to use for body text and for retained attributes

Method Summary

Modifier and Type	Method and Description
XmlTextAnnotation	createTextAnnotation(String xmlText, String corpusId, String docId) A method for creating TextAnnotation by tokenizing the given text string.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

XmlTextAnnotationMaker

public XmlTextAnnotationMaker(TextAnnotationBuilder taBuilder,
                              XmlDocumentProcessor xmlProcessor)

Specifies the behavior of the XmlTextAnnotationMaker: tokenization (via the TextAnnotationBuilder), which xml tags to use for body text and for retained attributes

Parameters:

taBuilder - generates the sentence split and tokenized text for further processing

xmlProcessor - responsible for parsing xml, extracting processable text and relevant markup info

Method Detail

createTextAnnotation

public XmlTextAnnotation createTextAnnotation(String xmlText,
                                              String corpusId,
                                              String docId)

A method for creating TextAnnotation by tokenizing the given text string.

Parameters:

xmlText - Raw xml text from corpus document

corpusId - corpus identifier

docId - text identifier

Returns:

an XmlTextAnnotation with the cleaned text (StringTransformation), TextAnnotation for the cleaned text, and xml markup extracted from source