org.knowceans.corpus.xpert
Class XptDb2Corpus

java.lang.Object
  org.knowceans.corpus.xpert.XptDb2Corpus

public final class XptDb2Corpus
extends java.lang.Object

Author:

Gregor Heinrich

Field Summary

private static java.util.HashMap<java.lang.Integer,XptAuthor>	allAuthors
private static java.util.Vector<XptDocument>	allDocs
private static IgdCategories	cat
private java.sql.Connection	con
private static XptDocument	curDoc
private static boolean	inSentence
private static java.io.Writer	out
private static java.lang.String	prevWord
private static Stemmer	stem
private static StopWordFilter	stop
boolean	useBigrams
boolean	useStemming

Constructor Summary

XptDb2Corpus(java.lang.String modelFile)
get properties for indexing and create database connection.

Method Summary

void	configure(boolean useStemming, boolean useBigrams) configure the db reader.
java.lang.String	convertToEntities(java.lang.String in)
java.lang.String	convertToUnicode(java.lang.String in)
XptAuthor	getAuthorData(int id) returns the list of authors.
java.util.Vector<java.lang.Integer>	getAuthorIds() gets all valid author ids from the database.
XptDocument	getProjectData(int pid) returns a single ProjectRec for the id.
java.util.Vector<java.lang.Integer>	getProjectIds() gets all valid project ids from the database.
static void	main(java.lang.String[] args)
private void	parseText(java.lang.String s, java.util.Vector<java.lang.String> words) Parse the given text and add terms to the model.
private void	read()
private java.lang.String	removePunct(java.lang.String s) Remove all punctuation
private AuthorTermCorpus	toCorpus()
private void	write(AuthorTermCorpus f, java.lang.String corpusname)

Methods inherited from class java.lang.Object

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

Field Detail

con

private java.sql.Connection con

out

private static java.io.Writer out

curDoc

private static XptDocument curDoc

inSentence

private static boolean inSentence

stop

private static StopWordFilter stop

useStemming

public boolean useStemming

useBigrams

public boolean useBigrams

prevWord

private static java.lang.String prevWord

cat

private static IgdCategories cat

stem

private static Stemmer stem

allDocs

private static java.util.Vector<XptDocument> allDocs

allAuthors

private static java.util.HashMap<java.lang.Integer,XptAuthor> allAuthors

Constructor Detail

XptDb2Corpus

public XptDb2Corpus(java.lang.String modelFile)

get properties for indexing and create database connection.

Method Detail

main

public static void main(java.lang.String[] args)

configure

public void configure(boolean useStemming,
                      boolean useBigrams)

configure the db reader.

Parameters:

useStemming - use stemming

useBigrams - use bigrams

read

private void read()

write

private void write(AuthorTermCorpus f,
                   java.lang.String corpusname)

toCorpus

private AuthorTermCorpus toCorpus()

getProjectIds

public java.util.Vector<java.lang.Integer> getProjectIds()

gets all valid project ids from the database.

Returns:

getAuthorIds

public java.util.Vector<java.lang.Integer> getAuthorIds()

gets all valid author ids from the database.

Returns:

getProjectData

public XptDocument getProjectData(int pid)

returns a single ProjectRec for the id.

Parameters:

pid -

Throws:

java.sql.SQLException

getAuthorData

public XptAuthor getAuthorData(int id)

returns the list of authors.

Parameters:

pid -

Throws:

java.sql.SQLException

parseText

private void parseText(java.lang.String s,
                       java.util.Vector<java.lang.String> words)

Parse the given text and add terms to the model. Here stop-words and stem filtering is located.

Parameters:

s -

removePunct

private java.lang.String removePunct(java.lang.String s)

Remove all punctuation

Parameters:

s -

Returns:

convertToEntities

public java.lang.String convertToEntities(java.lang.String in)

convertToUnicode

public java.lang.String convertToUnicode(java.lang.String in)