Package oxygen

Class OxygenMain

java.lang.Object

oxygen.OxygenMain

public class OxygenMain
extends java.lang.Object

Field Summary

Fields

Modifier and Type	Field	Description
private static java.lang.String	BODY_FIELD
private static java.lang.String	CATEGORY_FIELD
private static float	DEFAULT_LAMBDA
private static long	endIndexing
private static long	endQueryParse
private static long	endSearch
private static float	lambda
private static long	overallTime
private static java.lang.String	PATH_TO_ANSWERS_OUTPUT
private static java.lang.String	PATH_TO_INDEX1
private static java.lang.String	PATH_TO_INDEX2
private static java.lang.String	PATH_TO_JSON
private static java.lang.String	PATH_TO_QUESTIONS
private static java.lang.String	PATH_TO_QUESTIONS_MAIN
private static org.apache.lucene.search.similarities.Similarity[]	similarities
private static org.apache.lucene.search.similarities.MultiSimilarity	similarity
private static long	startIndexing
private static long	startQueryParse
private static long	startSearch
private static org.apache.lucene.document.FieldType	termVector_t

Constructor Summary

Constructors

Constructor	Description
OxygenMain()

Method Summary

Modifier and Type	Method	Description
private static void	appendToJson(QuestionAnswered q, java.io.FileWriter writer, boolean last)	Appends answers to json file, and a comma if the line is not last
private static float	calculateLambda()	Calculates lambda \lambda = 1 - \displaystyle\dfrac{1}{n}\sum_{q \in C_q}{\vert d \vert}
private static java.util.List<Answer>	createAnswersArray(org.apache.lucene.search.IndexSearcher searcher, org.apache.lucene.search.Query q, org.apache.lucene.search.TopDocs topDocs)
private static void	createQuestionsJson(java.lang.String filename)	Extracts questions from raw data and writes them to new json file
private static void	indexCorpus(org.apache.lucene.store.Directory dir, java.lang.String jsonPath, org.apache.lucene.analysis.Analyzer analyzer, org.apache.lucene.search.similarities.Similarity similarity)	Writes the corpus to given directory from given json file using given analyzer and similarity
static void	main(java.lang.String[] args)
private static org.apache.lucene.index.IndexWriterConfig	newIndexWriterConfig(org.apache.lucene.analysis.Analyzer analyzer, org.apache.lucene.search.similarities.Similarity similarity)	Creates index writer with provided analyzer and similarity
private static void	printCorpus(java.util.List<java.lang.String> corpus)
private static void	printQueryInfo(java.lang.String originalQuery, java.lang.String preFilteredQuery, org.apache.lucene.search.Query parsedQuery, long endQueryParseMillis, long startQueryParseMillis)	Outputs query breackdown to the screen
private static void	printSearchResults(org.apache.lucene.search.TopDocs td, org.apache.lucene.search.Query q, org.apache.lucene.index.DirectoryReader reader, org.apache.lucene.search.IndexSearcher searcher, long searchTimeMillis, long overallTimeMillis)
private static java.util.List<Answer>	search(org.apache.lucene.index.DirectoryReader reader, java.lang.String queryString, java.lang.String preFilteredQuery, OxygenAnalyzerBase analyzer)	Main search function

Methods inherited from class java.lang.Object

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

Field Detail

PATH_TO_JSON

private static final java.lang.String PATH_TO_JSON

See Also:

Constant Field Values

PATH_TO_INDEX1

private static final java.lang.String PATH_TO_INDEX1

See Also:

Constant Field Values

PATH_TO_INDEX2

private static final java.lang.String PATH_TO_INDEX2

See Also:

Constant Field Values

PATH_TO_QUESTIONS

private static final java.lang.String PATH_TO_QUESTIONS

See Also:

Constant Field Values

PATH_TO_QUESTIONS_MAIN

private static final java.lang.String PATH_TO_QUESTIONS_MAIN

See Also:

Constant Field Values

PATH_TO_ANSWERS_OUTPUT

private static final java.lang.String PATH_TO_ANSWERS_OUTPUT

See Also:

Constant Field Values

DEFAULT_LAMBDA

private static final float DEFAULT_LAMBDA

BODY_FIELD

private static final java.lang.String BODY_FIELD

See Also:

Constant Field Values

CATEGORY_FIELD

private static final java.lang.String CATEGORY_FIELD

See Also:

Constant Field Values

termVector_t

private static final org.apache.lucene.document.FieldType termVector_t

overallTime

private static long overallTime

startIndexing

private static long startIndexing

endIndexing

private static long endIndexing

startQueryParse

private static long startQueryParse

endQueryParse

private static long endQueryParse

startSearch

private static long startSearch

endSearch

private static long endSearch

lambda

private static float lambda

similarities

private static org.apache.lucene.search.similarities.Similarity[] similarities

similarity

private static org.apache.lucene.search.similarities.MultiSimilarity similarity

Constructor Detail

OxygenMain

public OxygenMain()

Method Detail

main

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

Throws:

java.lang.Exception

newIndexWriterConfig

private static org.apache.lucene.index.IndexWriterConfig newIndexWriterConfig(org.apache.lucene.analysis.Analyzer analyzer,
                                                                              org.apache.lucene.search.similarities.Similarity similarity)

Creates index writer with provided analyzer and similarity

Parameters:

analyzer - Analyzer used

similarity - Similarity used

Returns:

new IndexWriterConfing

indexCorpus

private static void indexCorpus(org.apache.lucene.store.Directory dir,
                                java.lang.String jsonPath,
                                org.apache.lucene.analysis.Analyzer analyzer,
                                org.apache.lucene.search.similarities.Similarity similarity)

Writes the corpus to given directory from given json file using given analyzer and similarity

Parameters:

dir -

jsonPath -

analyzer -

similarity -

search

private static java.util.List<Answer> search(org.apache.lucene.index.DirectoryReader reader,
                                             java.lang.String queryString,
                                             java.lang.String preFilteredQuery,
                                             OxygenAnalyzerBase analyzer)
                                      throws java.io.IOException,
                                             org.apache.lucene.queryparser.classic.ParseException,
                                             OxygenNotFound

Main search function

Parameters:

reader -

queryString -

preFilteredQuery -

analyzer -

Returns:

Throws:

java.io.IOException

org.apache.lucene.queryparser.classic.ParseException

OxygenNotFound

createAnswersArray

private static java.util.List<Answer> createAnswersArray(org.apache.lucene.search.IndexSearcher searcher,
                                                         org.apache.lucene.search.Query q,
                                                         org.apache.lucene.search.TopDocs topDocs)
                                                  throws java.io.IOException

Throws:

java.io.IOException

appendToJson

private static void appendToJson(QuestionAnswered q,
                                 java.io.FileWriter writer,
                                 boolean last)

Appends answers to json file, and a comma if the line is not last

Parameters:

q -

writer -

createQuestionsJson

private static void createQuestionsJson(java.lang.String filename)
                                 throws java.io.IOException

Extracts questions from raw data and writes them to new json file

Parameters:

filename -

Throws:

java.io.IOException

calculateLambda

private static float calculateLambda()

Calculates lambda \lambda = 1 - \displaystyle\dfrac{1}{n}\sum_{q \in C_q}{\vert d \vert}

Returns:

printCorpus

private static void printCorpus(java.util.List<java.lang.String> corpus)

printQueryInfo

private static void printQueryInfo(java.lang.String originalQuery,
                                   java.lang.String preFilteredQuery,
                                   org.apache.lucene.search.Query parsedQuery,
                                   long endQueryParseMillis,
                                   long startQueryParseMillis)

Outputs query breackdown to the screen

Parameters:

originalQuery -

preFilteredQuery -

parsedQuery -

endQueryParseMillis -

startQueryParseMillis -

printSearchResults

private static void printSearchResults(org.apache.lucene.search.TopDocs td,
                                       org.apache.lucene.search.Query q,
                                       org.apache.lucene.index.DirectoryReader reader,
                                       org.apache.lucene.search.IndexSearcher searcher,
                                       long searchTimeMillis,
                                       long overallTimeMillis)
                                throws java.io.IOException

Throws:

java.io.IOException