Package weka.core

Class DictionaryBuilder

  • All Implemented Interfaces:
    java.io.Serializable, Aggregateable<DictionaryBuilder>, OptionHandler
    public class DictionaryBuilder
extends java.lang.Object
implements Aggregateable<DictionaryBuilder>, OptionHandler, java.io.Serializable
    Class for building and maintaining a dictionary of terms. Has methods for loading, saving and aggregating dictionaries. Supports loading/saving in binary and textual format. Textual format is expected to have one or two comma separated values per line of the format. 

     term [,doc_count]
    where 
     doc_count
    is the number of documents that the term has occurred in.
    Version:
    $Revision: 14443 $
    Author:
    Mark Hall (mhall{[at]}pentaho{[dot]}com)
    See Also:
    Serialized Form

    • Constructor Detail

      • DictionaryBuilder

        public DictionaryBuilder()

    • Method Detail

      • setAverageDocLength

        @ProgrammaticProperty
public void setAverageDocLength​(double averageDocLength)
        Set the average document length to use when normalizing
        Parameters:
        averageDocLength - the average document length to use

      • getAverageDocLength

        public double getAverageDocLength()
        Get the average document length to use when normalizing
        Returns:
        the average document length

      • sortDictionaryTipText

        public java.lang.String sortDictionaryTipText()
        Tip text for this property
        Returns:
        the tip text for this property

      • setSortDictionary

        public void setSortDictionary​(boolean sortDictionary)
        Set whether to keep the dictionary sorted alphabetically as it is built. Setting this to true uses a TreeMap internally (which is slower than the default unsorted LinkedHashMap).
        Parameters:
        sortDictionary - true to keep the dictionary sorted alphabetically

      • getSortDictionary

        public boolean getSortDictionary()
        Get whether to keep the dictionary sorted alphabetically as it is built. Setting this to true uses a TreeMap internally (which is slower than the default unsorted LinkedHashMap).
        Returns:
        true to keep the dictionary sorted alphabetically

      • getOutputWordCounts

        public boolean getOutputWordCounts()
        Gets whether output instances contain 0 or 1 indicating word presence, or word counts.
        Returns:
        true if word counts should be output.

      • setOutputWordCounts

        public void setOutputWordCounts​(boolean outputWordCounts)
        Sets whether output instances contain 0 or 1 indicating word presence, or word counts.
        Parameters:
        outputWordCounts - true if word counts should be output.

      • outputWordCountsTipText

        public java.lang.String outputWordCountsTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getSelectedRange

        public Range getSelectedRange()
        Get the value of m_SelectedRange.
        Returns:
        Value of m_SelectedRange.

      • setSelectedRange

        public void setSelectedRange​(java.lang.String newSelectedRange)
        Set the value of m_SelectedRange.
        Parameters:
        newSelectedRange - Value to assign to m_SelectedRange.

      • attributeIndicesTipText

        public java.lang.String attributeIndicesTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getAttributeIndices

        public java.lang.String getAttributeIndices()
        Gets the current range selection.
        Returns:
        a string containing a comma separated list of ranges

      • setAttributeIndices

        public void setAttributeIndices​(java.lang.String rangeList)
        Sets which attributes are to be worked on.
        Parameters:
        rangeList - a string representing the list of attributes. Since the string will typically come from a user, attributes are indexed from 1.
        eg: first-3,5,6-last
        Throws:
        java.lang.IllegalArgumentException - if an invalid range list is supplied

      • setAttributeIndicesArray

        public void setAttributeIndicesArray​(int[] attributes)
        Sets which attributes are to be processed.
        Parameters:
        attributes - an array containing indexes of attributes to process. Since the array will typically come from a program, attributes are indexed from 0.
        Throws:
        java.lang.IllegalArgumentException - if an invalid set of ranges is supplied

      • invertSelectionTipText

        public java.lang.String invertSelectionTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getInvertSelection

        public boolean getInvertSelection()
        Gets whether the supplied columns are to be processed or skipped.
        Returns:
        true if the supplied columns will be kept

      • setInvertSelection

        public void setInvertSelection​(boolean invert)
        Sets whether selected columns should be processed or skipped.
        Parameters:
        invert - the new invert setting

      • getWordsToKeep

        public int getWordsToKeep()
        Gets the number of words (per class if there is a class attribute assigned) to attempt to keep.
        Returns:
        the target number of words in the output vector (per class if assigned).

      • setWordsToKeep

        public void setWordsToKeep​(int newWordsToKeep)
        Sets the number of words (per class if there is a class attribute assigned) to attempt to keep.
        Parameters:
        newWordsToKeep - the target number of words in the output vector (per class if assigned).

      • wordsToKeepTipText

        public java.lang.String wordsToKeepTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getPeriodicPruning

        public long getPeriodicPruning()
        Gets the rate (number of instances) at which the dictionary is periodically pruned.
        Returns:
        the rate at which the dictionary is periodically pruned

      • setPeriodicPruning

        public void setPeriodicPruning​(long newPeriodicPruning)
        Sets the rate (number of instances) at which the dictionary is periodically pruned
        Parameters:
        newPeriodicPruning - the rate at which the dictionary is periodically pruned

      • periodicPruningTipText

        public java.lang.String periodicPruningTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getTFTransform

        public boolean getTFTransform()
        Gets whether if the word frequencies should be transformed into log(1+fij) where fij is the frequency of word i in document(instance) j.
        Returns:
        true if word frequencies are to be transformed.

      • setTFTransform

        public void setTFTransform​(boolean TFTransform)
        Sets whether if the word frequencies should be transformed into log(1+fij) where fij is the frequency of word i in document(instance) j.
        Parameters:
        TFTransform - true if word frequencies are to be transformed.

      • TFTransformTipText

        public java.lang.String TFTransformTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getAttributeNamePrefix

        public java.lang.String getAttributeNamePrefix()
        Get the attribute name prefix.
        Returns:
        The current attribute name prefix.

      • setAttributeNamePrefix

        public void setAttributeNamePrefix​(java.lang.String newPrefix)
        Set the attribute name prefix.
        Parameters:
        newPrefix - String to use as the attribute name prefix.

      • attributeNamePrefixTipText

        public java.lang.String attributeNamePrefixTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getIDFTransform

        public boolean getIDFTransform()
        Sets whether if the word frequencies in a document should be transformed into:
        fij*log(num of Docs/num of Docs with word i)
        where fij is the frequency of word i in document(instance) j.
        Returns:
        true if the word frequencies are to be transformed.

      • setIDFTransform

        public void setIDFTransform​(boolean IDFTransform)
        Sets whether if the word frequencies in a document should be transformed into:
        fij*log(num of Docs/num of Docs with word i)
        where fij is the frequency of word i in document(instance) j.
        Parameters:
        IDFTransform - true if the word frequecies are to be transformed

      • IDFTransformTipText

        public java.lang.String IDFTransformTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getNormalize

        public boolean getNormalize()
        Get whether word frequencies for a document should be normalized
        Returns:
        true if word frequencies should be normalized

      • setNormalize

        public void setNormalize​(boolean n)
        Set whether word frequencies for a document should be normalized
        Parameters:
        n - true if word frequencies should be normalized

      • normalizeTipText

        public java.lang.String normalizeTipText()
        Tip text for this property
        Returns:
        the tip text for this property

      • normalizeDocLengthTipText

        public java.lang.String normalizeDocLengthTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getLowerCaseTokens

        public boolean getLowerCaseTokens()
        Gets whether if the tokens are to be downcased or not.
        Returns:
        true if the tokens are to be downcased.

      • setLowerCaseTokens

        public void setLowerCaseTokens​(boolean downCaseTokens)
        Sets whether if the tokens are to be downcased or not. (Doesn't affect non-alphabetic characters in tokens).
        Parameters:
        downCaseTokens - should be true if only lower case tokens are to be formed.

      • lowerCaseTokensTipText

        public java.lang.String lowerCaseTokensTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • doNotOperateOnPerClassBasisTipText

        public java.lang.String doNotOperateOnPerClassBasisTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getDoNotOperateOnPerClassBasis

        public boolean getDoNotOperateOnPerClassBasis()
        Get the DoNotOperateOnPerClassBasis value.
        Returns:
        the DoNotOperateOnPerClassBasis value.

      • setDoNotOperateOnPerClassBasis

        public void setDoNotOperateOnPerClassBasis​(boolean newDoNotOperateOnPerClassBasis)
        Set the DoNotOperateOnPerClassBasis value.
        Parameters:
        newDoNotOperateOnPerClassBasis - The new DoNotOperateOnPerClassBasis value.

      • minTermFreqTipText

        public java.lang.String minTermFreqTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getMinTermFreq

        public int getMinTermFreq()
        Get the MinTermFreq value.
        Returns:
        the MinTermFreq value.

      • setMinTermFreq

        public void setMinTermFreq​(int newMinTermFreq)
        Set the MinTermFreq value.
        Parameters:
        newMinTermFreq - The new MinTermFreq value.

      • getStemmer

        public Stemmer getStemmer()
        Returns the current stemming algorithm, null if none is used.
        Returns:
        the current stemming algorithm, null if none set

      • setStemmer

        public void setStemmer​(Stemmer value)
        the stemming algorithm to use, null means no stemming at all (i.e., the NullStemmer is used).
        Parameters:
        value - the configured stemming algorithm, or null
        See Also:
        NullStemmer

      • stemmerTipText

        public java.lang.String stemmerTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getStopwordsHandler

        public StopwordsHandler getStopwordsHandler()
        Gets the stopwords handler.
        Returns:
        the stopwords handler

      • setStopwordsHandler

        public void setStopwordsHandler​(StopwordsHandler value)
        Sets the stopwords handler to use.
        Parameters:
        value - the stopwords handler, if null, Null is used

      • stopwordsHandlerTipText

        public java.lang.String stopwordsHandlerTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • getTokenizer

        public Tokenizer getTokenizer()
        Returns the current tokenizer algorithm.
        Returns:
        the current tokenizer algorithm

      • setTokenizer

        public void setTokenizer​(Tokenizer value)
        the tokenizer algorithm to use.
        Parameters:
        value - the configured tokenizing algorithm

      • tokenizerTipText

        public java.lang.String tokenizerTipText()
        Returns the tip text for this property.
        Returns:
        tip text for this property suitable for displaying in the explorer/experimenter gui

      • listOptions

        public java.util.Enumeration<Option> listOptions()
        Returns an enumeration describing the available options.
        Specified by:
        listOptions in interface OptionHandler
        Returns:
        an enumeration of all the available options

      • getOptions

        public java.lang.String[] getOptions()
        Gets the current settings of the DictionaryBuilder
        Specified by:
        getOptions in interface OptionHandler
        Returns:
        an array of strings suitable for passing to setOptions

      • setOptions

        public void setOptions​(java.lang.String[] options)
                throws java.lang.Exception
        Parses a given list of options.

        Valid options are: 

         -C
  Output word counts rather than boolean word presence.
 
         -R <index1,index2-index4,...>
  Specify list of string attributes to convert to words (as weka Range).
  (default: select all string attributes)
 
         -V
  Invert matching sense of column indexes.
 
         -P <attribute name prefix>
  Specify a prefix for the created attribute names.
  (default: "")
 
         -W <number of words to keep>
  Specify approximate number of word fields to create.
  Surplus words will be discarded..
  (default: 1000)
 
         -prune-rate <rate as a percentage of dataset>
  Specify the rate (e.g., every 10% of the input dataset) at which to periodically prune the dictionary.
  -W prunes after creating a full dictionary. You may not have enough memory for this approach.
  (default: no periodic pruning)
 
         -T
  Transform the word frequencies into log(1+fij)
  where fij is the frequency of word i in jth document(instance).
 
         -I
  Transform each word frequency into:
  fij*log(num of Documents/num of documents containing word i)
    where fij if frequency of word i in jth document(instance)
 
         -N
  Whether to 0=not normalize/1=normalize all data/2=normalize test data only
  to average length of training documents (default 0=don't normalize).
 
         -L
  Convert all tokens to lowercase before adding to the dictionary.
 
         -stopwords-handler
  The stopwords handler to use (default Null).
 
         -stemmer <spec>
  The stemming algorithm (classname plus parameters) to use.
 
         -M <int>
  The minimum term frequency (default = 1).
 
         -O
  If this is set, the maximum number of words and the
  minimum term frequency is not enforced on a per-class
  basis but based on the documents in all the classes
  (even if a class attribute is set).
 
         -tokenizer <spec>
  The tokenizing algorihtm (classname plus parameters) to use.
  (default: weka.core.tokenizers.WordTokenizer)
        Specified by:
        setOptions in interface OptionHandler
        Parameters:
        options - the list of options as an array of strings
        Throws:
        java.lang.Exception - if an option is not supported

      • setup

        public void setup​(Instances inputFormat)
           throws java.lang.Exception
        Throws:
        java.lang.Exception

      • getInputFormat

        public Instances getInputFormat()
        Gets the currently set input format
        Returns:
        the current input format

      • readyToVectorize

        public boolean readyToVectorize()
        Returns true if this DictionaryBuilder is ready to vectorize incoming instances
        Returns:
        true if we can vectorize incoming instances

      • getVectorizedFormat

        public Instances getVectorizedFormat()
                              throws java.lang.Exception
        Get the output format
        Returns:
        the output format
        Throws:
        java.lang.Exception - if there is no input format set and/or the dictionary has not been constructed yet.

      • vectorizeBatch

        public Instances vectorizeBatch​(Instances batch,
                                boolean setAvgDocLength)
                         throws java.lang.Exception
        Convert a batch of instances
        Parameters:
        batch - the batch to convert.
        setAvgDocLength - true to compute and set the average document length for this DictionaryBuilder from the batch - this uses the final pruned dictionary when computing doc lengths. When vectorizing non-training batches, and normalization has been turned on, this should be set to false.
        Returns:
        the converted batch
        Throws:
        java.lang.Exception - if there is no input format set and/or the dictionary has not been constructed yet.

      • vectorizeInstance

        public Instance vectorizeInstance​(Instance input)
                           throws java.lang.Exception
        Convert an input instance. Any string attributes not being vectorized do not have their values retained in memory (i.e. only the string values for the instance being vectorized are held in memory).
        Parameters:
        input - the input instance
        Returns:
        a converted instance
        Throws:
        java.lang.Exception - if there is no input format set and/or the dictionary has not been constructed yet.

      • vectorizeInstance

        public Instance vectorizeInstance​(Instance input,
                                  boolean retainStringAttValuesInMemory)
                           throws java.lang.Exception
        Convert an input instance.
        Parameters:
        input - the input instance
        retainStringAttValuesInMemory - true if the values of string attributes not being vectorized should be retained in memory
        Returns:
        a converted instance
        Throws:
        java.lang.Exception - if there is no input format set and/or the dictionary has not been constructed yet

      • processInstance

        public void processInstance​(Instance inst)
        Process an instance by tokenizing string attributes and updating the dictionary.
        Parameters:
        inst - the instance to process

      • reset

        public void reset()
        Clear the dictionary(s)

      • getDictionaries

        public java.util.Map<java.lang.String,​int[]>[] getDictionaries​(boolean minFrequencyPrune)
                                                              throws WekaException
        Get the current dictionary(s) (one per class for nominal class, if set). These are the dictionaries that are built/updated when processInstance() is called. The finalized dictionary (used for vectorization) can be obtained by calling finalizeDictionary() - this returns a consolidated (over classes) and pruned final dictionary.
        Parameters:
        minFrequencyPrune - prune the dictionaries of low frequency terms before returning them
        Returns:
        the dictionaries
        Throws:
        WekaException

      • finalizeAggregation

        public void finalizeAggregation()
                         throws java.lang.Exception
        Description copied from interface: Aggregateable
        Call to complete the aggregation process. Allows implementers to do any final processing based on how many objects were aggregated.
        Specified by:
        finalizeAggregation in interface Aggregateable<DictionaryBuilder>
        Throws:
        java.lang.Exception - if the aggregation can't be finalized for some reason

      • finalizeDictionary

        public java.util.Map<java.lang.String,​int[]> finalizeDictionary()
                                                               throws java.lang.Exception
        Performs final pruning and consolidation according to the number of words to keep property. Finalization is performed just once, subsequent calls to this method return the finalized dictionary computed on the first call (unless reset() has been called in between).
        Returns:
        the consolidated and pruned final dictionary, or null if the input format did not contain any string attributes within the selected range to process
        Throws:
        java.lang.Exception - if a problem occurs

      • loadDictionary

        public void loadDictionary​(java.lang.String filename,
                           boolean plainText)
                    throws java.io.IOException
        Load a dictionary from a file
        Parameters:
        filename - the file to load from
        plainText - true if the dictionary is in text format
        Throws:
        java.io.IOException - if a problem occurs

      • loadDictionary

        public void loadDictionary​(java.io.File toLoad,
                           boolean plainText)
                    throws java.io.IOException
        Load a dictionary from a file
        Parameters:
        toLoad - the file to load from
        plainText - true if the dictionary is in text format
        Throws:
        java.io.IOException - if a problem occurs

      • loadDictionary

        public void loadDictionary​(java.io.Reader reader)
                    throws java.io.IOException
        Load a textual dictionary from a reader
        Parameters:
        reader - the reader to read from
        Throws:
        java.io.IOException - if a problem occurs

      • loadDictionary

        public void loadDictionary​(java.io.InputStream is)
                    throws java.io.IOException
        Load a binary dictionary from an input stream
        Parameters:
        is - the input stream to read from
        Throws:
        java.io.IOException - if a problem occurs

      • saveDictionary

        public void saveDictionary​(java.lang.String filename,
                           boolean plainText)
                    throws java.io.IOException
        Save the dictionary
        Parameters:
        filename - the file to save to
        plainText - true if the dictionary should be saved in text format
        Throws:
        java.io.IOException - if a problem occurs

      • saveDictionary

        public void saveDictionary​(java.io.File toSave,
                           boolean plainText)
                    throws java.io.IOException
        Save a dictionary
        Parameters:
        toSave - the file to save to
        plainText - true if the dictionary should be saved in text format
        Throws:
        java.io.IOException - if a problem occurs

      • saveDictionary

        public void saveDictionary​(java.io.Writer writer)
                    throws java.io.IOException
        Save the dictionary in textual format
        Parameters:
        writer - the writer to write to
        Throws:
        java.io.IOException - if a problem occurs

      • saveDictionary

        public void saveDictionary​(java.io.OutputStream os)
                    throws java.io.IOException
        Save the dictionary in binary form
        Parameters:
        os - the output stream to write to
        Throws:
        java.io.IOException - if a problem occurs