attribute.java

:<<数据挖掘--实用机器学习技术及java实现>>一书的配套源程序

      } else {
	text.append("string");
      }
    }
    return text.toString();
  }

  /**
   * Returns the attribute's type as an integer.
   *
   * @returns the attribute's type.
   */
  public final int type() {

    return m_Type;
  }

  /**
   * Returns a value of a nominal or string attribute. 
   * Returns an empty string if the attribute is neither
   * nominal nor a string attribute.
   *
   * @param valIndex the value's index
   * @return the attribute's value as a string
   */
  public final String value(int valIndex) {
    
    if (!isNominal() && !isString()) {
      return "";
    } else {
      Object val = m_Values.elementAt(valIndex);
      
      // If we're storing strings compressed, uncompress it.
      if (val instanceof SerializedObject) {
        val = ((SerializedObject)val).getObject();
      }
      return (String) val;
    }
  }

  /**
   * Constructor for a numeric attribute with a particular index.
   *
   * @param attributeName the name for the attribute
   * @param index the attribute's index
   */
  Attribute(String attributeName, int index) {

    this(attributeName);

    m_Index = index;
  }

  /**
   * Constructor for nominal attributes and string attributes with
   * a particular index.
   * If a null vector of attribute values is passed to the method,
   * the attribute is assumed to be a string.
   *
   * @param attributeName the name for the attribute
   * @param attributeValues a vector of strings denoting the attribute values.
   * Null if the attribute is a string attribute.
   * @param index the attribute's index
   */
  Attribute(String attributeName, FastVector attributeValues, 
	    int index) {

    this(attributeName, attributeValues);

    m_Index = index;
  }

  /**
   * Adds a string value to the list of valid strings for attributes
   * of type STRING and returns the index of the string.
   *
   * @param value The string value to add
   * @return the index assigned to the string, or -1 if the attribute is not
   * of type Attribute.STRING 
   */
  public int addStringValue(String value) {

    if (!isString()) {
      return -1;
    }
    Object store = value;

    if (value.length() > STRING_COMPRESS_THRESHOLD) {
      try {
        store = new SerializedObject(value, true);
      } catch (Exception ex) {
        System.err.println("Couldn't compress string attribute value -"
                           + " storing uncompressed.");
      }
    }
    Integer index = (Integer)m_Hashtable.get(store);
    if (index != null) {
      return index.intValue();
    } else {
      int intIndex = m_Values.size();
      m_Values.addElement(store);
      m_Hashtable.put(store, new Integer(intIndex));
      return intIndex;
    }
  }

  /**
   * Adds a string value to the list of valid strings for attributes
   * of type STRING and returns the index of the string. This method is
   * more efficient than addStringValue(String) for long strings.
   *
   * @param src The Attribute containing the string value to add.
   * @param int index the index of the string value in the source attribute.
   * @return the index assigned to the string, or -1 if the attribute is not
   * of type Attribute.STRING 
   */
  public int addStringValue(Attribute src, int index) {

    if (!isString()) {
      return -1;
    }
    Object store = src.m_Values.elementAt(index);
    Integer oldIndex = (Integer)m_Hashtable.get(store);
    if (oldIndex != null) {
      return oldIndex.intValue();
    } else {
      int intIndex = m_Values.size();
      m_Values.addElement(store);
      m_Hashtable.put(store, new Integer(intIndex));
      return intIndex;
    }
  }

  /**
   * Adds an attribute value. Creates a fresh list of attribute
   * values before adding it.
   *
   * @param value the attribute value
   */
  final void addValue(String value) {

    m_Values = (FastVector)m_Values.copy();
    m_Hashtable = (Hashtable)m_Hashtable.clone();
    forceAddValue(value);
  }

  /**
   * Produces a shallow copy of this attribute with a new name.
   *
   * @param newName the name of the new attribute
   * @return a copy of this attribute with the same index
   */
  final Attribute copy(String newName) {

    Attribute copy = new Attribute(newName);

    copy.m_Index = m_Index;
    if (!isNominal() && !isString())
      return copy;
    copy.m_Type = m_Type;
    copy.m_Values = m_Values;
    copy.m_Hashtable = m_Hashtable;
 
    return copy;
  }

  /**
   * Removes a value of a nominal or string attribute. Creates a 
   * fresh list of attribute values before removing it.
   *
   * @param index the value's index
   * @exception IllegalArgumentException if the attribute is not nominal
   */
  final void delete(int index) {
    
    if (!isNominal() && !isString()) 
      throw new IllegalArgumentException("Can only remove value of" +
                                         "nominal or string attribute!");
    else {
      m_Values = (FastVector)m_Values.copy();
      m_Values.removeElementAt(index);
      Hashtable hash = new Hashtable(m_Hashtable.size());
      Enumeration enum = m_Hashtable.keys();
      while (enum.hasMoreElements()) {
	Object string = enum.nextElement();
	Integer valIndexObject = (Integer)m_Hashtable.get(string);
	int valIndex = valIndexObject.intValue();
	if (valIndex > index) {
	  hash.put(string, new Integer(valIndex - 1));
	} else if (valIndex < index) {
	  hash.put(string, valIndexObject);
	}
      }
      m_Hashtable = hash;
    }
  }

  /**
   * Adds an attribute value.
   *
   * @param value the attribute value
   */
  final void forceAddValue(String value) {

    Object store = value;
    if (value.length() > STRING_COMPRESS_THRESHOLD) {
      try {
        store = new SerializedObject(value, true);
      } catch (Exception ex) {
        System.err.println("Couldn't compress string attribute value -"
                           + " storing uncompressed.");
      }
    }
    m_Values.addElement(store);
    m_Hashtable.put(store, new Integer(m_Values.size() - 1));
  }

  /**
   * Sets the index of this attribute.
   *
   * @param the index of this attribute
   */
  final void setIndex(int index) {

    m_Index = index;
  }

  /**
   * Sets a value of a nominal attribute or string attribute.
   * Creates a fresh list of attribute values before it is set.
   *
   * @param index the value's index
   * @param string the value
   * @exception IllegalArgumentException if the attribute is not nominal or 
   * string.
   */
  final void setValue(int index, String string) {

    if (!isNominal() && !isString()) {
      throw new IllegalArgumentException("Can only set value of nominal"+
                                         "or string attribute!");
    } else {
      m_Values = (FastVector)m_Values.copy();
      m_Hashtable = (Hashtable)m_Hashtable.clone();
      Object store = string;
      if (string.length() > STRING_COMPRESS_THRESHOLD) {
        try {
          store = new SerializedObject(string, true);
        } catch (Exception ex) {
          System.err.println("Couldn't compress string attribute value -"
                             + " storing uncompressed.");
        }
      }
      m_Hashtable.remove(m_Values.elementAt(index));
      m_Values.setElementAt(store, index);
      m_Hashtable.put(store, new Integer(index));
    }
  }

  /**
   * Simple main method for testing this class.
   */
  public static void main(String[] ops) {

    try {
      
      // Create numeric attributes "length" and "weight"
      Attribute length = new Attribute("length");
      Attribute weight = new Attribute("weight");
      
      // Create vector to hold nominal values "first", "second", "third" 
      FastVector my_nominal_values = new FastVector(3); 
      my_nominal_values.addElement("first"); 
      my_nominal_values.addElement("second"); 
      my_nominal_values.addElement("third"); 
      
      // Create nominal attribute "position" 
      Attribute position = new Attribute("position", my_nominal_values);

      // Print the name of "position"
      System.out.println("Name of \"position\": " + position.name());

      // Print the values of "position"
      Enumeration attValues = position.enumerateValues();
      while (attValues.hasMoreElements()) {
	String string = (String)attValues.nextElement();
	System.out.println("Value of \"position\": " + string);
      }

      // Shallow copy attribute "position"
      Attribute copy = (Attribute) position.copy();

      // Test if attributes are the same
      System.out.println("Copy is the same as original: " + copy.equals(position));

      // Print index of attribute "weight" (should be unset: -1)
      System.out.println("Index of attribute \"weight\" (should be -1): " + 
			 weight.index());

      // Print index of value "first" of attribute "position"
      System.out.println("Index of value \"first\" of \"position\" (should be 0): " +
			 position.indexOfValue("first"));

      // Tests type of attribute "position"
      System.out.println("\"position\" is numeric: " + position.isNumeric());
      System.out.println("\"position\" is nominal: " + position.isNominal());
      System.out.println("\"position\" is string: " + position.isString());

      // Prints name of attribute "position"
      System.out.println("Name of \"position\": " + position.name());
    
      // Prints number of values of attribute "position"
      System.out.println("Number of values for \"position\": " + position.numValues());

      // Prints the values (againg)
      for (int i = 0; i < position.numValues(); i++) {
	System.out.println("Value " + i + ": " + position.value(i));
      }

      // Prints the attribute "position" in ARFF format
      System.out.println(position);

      // Checks type of attribute "position" using constants
      switch (position.type()) {
      case Attribute.NUMERIC:
	System.out.println("\"position\" is numeric");
	break;
      case Attribute.NOMINAL:
	System.out.println("\"position\" is nominal");
	break;
      case Attribute.STRING:
	System.out.println("\"position\" is string");
	break;
      default:
	System.out.println("\"position\" has unknown type");
      }
    } catch (Exception e) {
      e.printStackTrace();
    }
  }
}