Package weka.associations

Class LabeledItemSet

java.lang.Object
weka.associations.ItemSet
weka.associations.LabeledItemSet
All Implemented Interfaces:
Serializable, RevisionHandler
public class LabeledItemSet extends ItemSet implements Serializable, RevisionHandler
Class for storing a set of items together with a class label. Item sets are stored in a lexicographic order, which is determined by the header information of the set of instances used for generating the set of items. All methods in this class assume that item sets are stored in lexicographic order. The class provides the methods used for item sets in class association rule mining. Because every item set knows its class label the training set can be splitted up virtually.
Version:
$Revision: 10203 $
Author:
Stefan Mutter (mutter@cs.waikato.ac.nz)
See Also:

  • Constructor Details

    • LabeledItemSet

      public LabeledItemSet(int totalTrans, int classLabel)
      Constructor
      Parameters:
      totalTrans - the total number of transactions
      classLabel - the class lebel

  • Method Details

    • deleteItemSets

      public static ArrayList<Object> deleteItemSets(ArrayList<Object> itemSets, int minSupport, int maxSupport)
      Deletes all item sets that don't have minimum support and have more than maximum support
      Parameters:
      maxSupport - the maximum support
      itemSets - the set of item sets to be pruned
      minSupport - the minimum number of transactions to be covered
      Returns:
      the reduced set of item sets

    • equals

      public final boolean equals(Object itemSet)
      Tests if two item sets are equal.
      Overrides:
      equals in class ItemSet
      Parameters:
      itemSet - another item set
      Returns:
      true if this item set contains the same items as the given one

    • equalCondset

      public final boolean equalCondset(Object itemSet)
      Compares two item sets
      Parameters:
      itemSet - an item set
      Returns:
      true if the the item sets are equal, false otherwise

    • getHashtable

      public static Hashtable<ItemSet,Integer> getHashtable(ArrayList<Object> itemSets, int initialSize)
      Return a hashtable filled with the given item sets.
      Parameters:
      itemSets - the set of item sets to be used for filling the hash table
      initialSize - the initial size of the hashtable
      Returns:
      the generated hashtable

    • mergeAllItemSets

      public static ArrayList<Object> mergeAllItemSets(ArrayList<Object> itemSets, int size, int totalTrans)
      Merges all item sets in the set of (k-1)-item sets to create the (k)-item sets and updates the counters.
      Parameters:
      totalTrans - the total number of transactions
      itemSets - the set of (k-1)-item sets
      size - the value of (k-1)
      Returns:
      the generated (k)-item sets

    • divide

      public static Instances divide(Instances instances, boolean invert) throws Exception
      Splits the class attribute away. Depending on the invert flag, the instances without class attribute or only the class attribute of all instances is returned
      Parameters:
      instances - the instances
      invert - flag; if true only the class attribute remains, otherweise the class attribute is the only attribute that is deleted.
      Returns:
      Instances without the class attribute or instances with only the class attribute
      Throws:
      Exception - exception if instances cannot be splitted

    • singletons

      public static ArrayList<Object> singletons(Instances instancesNoClass, Instances classes) throws Exception
      Converts the header info of the given set of instances into a set of item sets (singletons). The ordering of values in the header file determines the lexicographic order. Each item set knows its class label.
      Parameters:
      instancesNoClass - instances without the class attribute
      classes - the values of the class attribute sorted according to instances
      Returns:
      a set of item sets, each containing a single item
      Throws:
      Exception - if singletons can't be generated successfully

    • pruneItemSets

      public static ArrayList<Object> pruneItemSets(ArrayList<Object> toPrune, Hashtable<ItemSet,Integer> kMinusOne)
      Prunes a set of (k)-item sets using the given (k-1)-item sets.
      Parameters:
      toPrune - the set of (k)-item sets to be pruned
      kMinusOne - the (k-1)-item sets to be used for pruning
      Returns:
      the pruned set of item sets

    • support

      public final int support()
      Outputs the support for an item set.
      Overrides:
      support in class ItemSet
      Returns:
      the support

    • upDateCounter

      public final void upDateCounter(Instance instanceNoClass, Instance instanceClass)
      Updates counter of item set with respect to given transaction.
      Parameters:
      instanceNoClass - instances without the class attribute
      instanceClass - the values of the class attribute sorted according to instances

    • upDateCounterTreatZeroAsMissing

      public final void upDateCounterTreatZeroAsMissing(Instance instanceNoClass, Instance instanceClass)
      Updates counter of item set with respect to given transaction.
      Parameters:
      instanceNoClass - instances without the class attribute
      instanceClass - the values of the class attribute sorted according to instances

    • upDateCounters

      public static void upDateCounters(ArrayList<Object> itemSets, Instances instancesNoClass, Instances instancesClass)
      Updates counter of a specific item set
      Parameters:
      itemSets - an item sets
      instancesNoClass - instances without the class attribute
      instancesClass - the values of the class attribute sorted according to instances

    • upDateCountersTreatZeroAsMissing

      public static void upDateCountersTreatZeroAsMissing(ArrayList<LabeledItemSet> itemSets, Instances instancesNoClass, Instances instancesClass)
      Updates counter of a specific item set
      Parameters:
      itemSets - an item sets
      instancesNoClass - instances without the class attribute
      instancesClass - the values of the class attribute sorted according to instances

    • generateRules

      public final ArrayList<Object>[] generateRules(double minConfidence, boolean noPrune)
      Generates rules out of item sets
      Parameters:
      minConfidence - the minimum confidence
      noPrune - flag indicating whether the rules are pruned accoridng to the minimum confidence value
      Returns:
      a set of rules

    • getRevision

      public String getRevision()
      Returns the revision string.
      Specified by:
      getRevision in interface RevisionHandler
      Overrides:
      getRevision in class ItemSet
      Returns:
      the revision