ir.vsr
Class Feedback

java.lang.Object
  ir.vsr.Feedback

public class Feedback
extends java.lang.Object

Gets and stores information about relevance feedback from the user and computes an updated query based on original query and retrieved documents that are rated relevant and irrelevant.

Field Summary

static double	ALPHA A Rochio/Ide algorithm parameter
java.util.ArrayList<DocumentReference>	badDocRefs The list of DocumentReference's that were rated irrelevant
static double	BETA A Rochio/Ide algorithm parameter
static double	GAMMA A Rochio/Ide algorithm parameter
java.util.ArrayList<DocumentReference>	goodDocRefs The list of DocumentReference's that were rated relevant
InvertedIndex	invertedIndex The current InvertedIndex
HashMapVector	queryVector The original query vector for this query
Retrieval[]	retrievals The current list of ranked retrievals

Constructor Summary

Feedback(HashMapVector queryVector, Retrieval[] retrievals, InvertedIndex invertedIndex)
Create a feedback object for this query with initial retrievals to be rated

Method Summary

void	addBad(DocumentReference docRef) Add a document to the list of those deemed irrelevant
void	addGood(DocumentReference docRef) Add a document to the list of those deemed relevant
void	getFeedback(int showNumber) Prompt the user for feedback on this numbered retrieval
boolean	haveFeedback(int showNumber) Has the user already provided feedback on this numbered retrieval?
boolean	isEmpty() Has the user rated any documents yet?
HashMapVector	newQuery() Use the Ide_regular algorithm to compute a new revised query.

Methods inherited from class java.lang.Object

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

Field Detail

ALPHA

public static double ALPHA

A Rochio/Ide algorithm parameter

BETA

public static double BETA

A Rochio/Ide algorithm parameter

GAMMA

public static double GAMMA

A Rochio/Ide algorithm parameter

queryVector

public HashMapVector queryVector

The original query vector for this query

retrievals

public Retrieval[] retrievals

The current list of ranked retrievals

invertedIndex

public InvertedIndex invertedIndex

The current InvertedIndex

goodDocRefs

public java.util.ArrayList<DocumentReference> goodDocRefs

The list of DocumentReference's that were rated relevant

badDocRefs

public java.util.ArrayList<DocumentReference> badDocRefs

The list of DocumentReference's that were rated irrelevant

Constructor Detail

Feedback

public Feedback(HashMapVector queryVector,
                Retrieval[] retrievals,
                InvertedIndex invertedIndex)

Create a feedback object for this query with initial retrievals to be rated

Method Detail

addGood

public void addGood(DocumentReference docRef)

Add a document to the list of those deemed relevant

addBad

public void addBad(DocumentReference docRef)

Add a document to the list of those deemed irrelevant

isEmpty

public boolean isEmpty()

Has the user rated any documents yet?

getFeedback

public void getFeedback(int showNumber)

Prompt the user for feedback on this numbered retrieval

haveFeedback

public boolean haveFeedback(int showNumber)

Has the user already provided feedback on this numbered retrieval?

newQuery

public HashMapVector newQuery()

Use the Ide_regular algorithm to compute a new revised query.

Returns:

The revised query vector.