dbscan.java

数据挖掘中聚类的算法

                        1,
                        "-I <String>"));
        vector.addElement(
                new Option("\tdistance-type (default = weka.clusterers.forOPTICSAndDBScan.DataObjects.EuclidianDataObject)",
                        "D",
                        1,
                        "-D <String>"));
        return vector.elements();
    }

    /**
     * Sets the OptionHandler's options using the given list. All options
     * will be set (or reset) during this call (i.e. incremental setting
     * of options is not possible). <p/>
     *
     <!-- options-start -->
     * Valid options are: <p/>
     * 
     * <pre> -E &lt;double&gt;
     *  epsilon (default = 0.9)</pre>
     * 
     * <pre> -M &lt;int&gt;
     *  minPoints (default = 6)</pre>
     * 
     * <pre> -I &lt;String&gt;
     *  index (database) used for DBScan (default = weka.clusterers.forOPTICSAndDBScan.Databases.SequentialDatabase)</pre>
     * 
     * <pre> -D &lt;String&gt;
     *  distance-type (default = weka.clusterers.forOPTICSAndDBScan.DataObjects.EuclidianDataObject)</pre>
     * 
     <!-- options-end -->
     *
     * @param options The list of options as an array of strings
     * @throws java.lang.Exception If an option is not supported
     */
    public void setOptions(String[] options) throws Exception {
        String optionString = Utils.getOption('E', options);
        if (optionString.length() != 0) {
            setEpsilon(Double.parseDouble(optionString));
        }

        optionString = Utils.getOption('M', options);
        if (optionString.length() != 0) {
            setMinPoints(Integer.parseInt(optionString));
        }

        optionString = Utils.getOption('I', options);
        if (optionString.length() != 0) {
            setDatabase_Type(optionString);
        }

        optionString = Utils.getOption('D', options);
        if (optionString.length() != 0) {
            setDatabase_distanceType(optionString);
        }
    }

    /**
     * Gets the current option settings for the OptionHandler.
     *
     * @return String[] The list of current option settings as an array of strings
     */
    public String[] getOptions() {
        String[] options = new String[8];
        int current = 0;

        options[current++] = "-E";
        options[current++] = "" + getEpsilon();
        options[current++] = "-M";
        options[current++] = "" + getMinPoints();
        options[current++] = "-I";
        options[current++] = "" + getDatabase_Type();
        options[current++] = "-D";
        options[current++] = "" + getDatabase_distanceType();

        return options;
    }

    /**
     * Returns a new Class-Instance of the specified database
     * @param database_Type String of the specified database
     * @param instances Instances that were delivered from WEKA
     * @return Database New constructed Database
     */
    public Database databaseForName(String database_Type, Instances instances) {
        Object o = null;

        Constructor co = null;
        try {
            co = (Class.forName(database_Type)).getConstructor(new Class[]{Instances.class});
            o = co.newInstance(new Object[]{instances});
        } catch (NoSuchMethodException e) {
            e.printStackTrace();
        } catch (SecurityException e) {
            e.printStackTrace();
        } catch (ClassNotFoundException e) {
            e.printStackTrace();
        } catch (InstantiationException e) {
            e.printStackTrace();
        } catch (IllegalAccessException e) {
            e.printStackTrace();
        } catch (InvocationTargetException e) {
            e.printStackTrace();
        }

        return (Database) o;
    }

    /**
     * Returns a new Class-Instance of the specified database
     * @param database_distanceType String of the specified distance-type
     * @param instance The original instance that needs to hold by this DataObject
     * @param key Key for this DataObject
     * @param database Link to the database
     * @return DataObject New constructed DataObject
     */
    public DataObject dataObjectForName(String database_distanceType, Instance instance, String key, Database database) {
        Object o = null;

        Constructor co = null;
        try {
            co = (Class.forName(database_distanceType)).
                    getConstructor(new Class[]{Instance.class, String.class, Database.class});
            o = co.newInstance(new Object[]{instance, key, database});
        } catch (NoSuchMethodException e) {
            e.printStackTrace();
        } catch (SecurityException e) {
            e.printStackTrace();
        } catch (ClassNotFoundException e) {
            e.printStackTrace();
        } catch (InstantiationException e) {
            e.printStackTrace();
        } catch (IllegalAccessException e) {
            e.printStackTrace();
        } catch (InvocationTargetException e) {
            e.printStackTrace();
        }

        return (DataObject) o;
    }

    /**
     * Sets a new value for minPoints
     * @param minPoints MinPoints
     */
    public void setMinPoints(int minPoints) {
        this.minPoints = minPoints;
    }

    /**
     * Sets a new value for epsilon
     * @param epsilon Epsilon
     */
    public void setEpsilon(double epsilon) {
        this.epsilon = epsilon;
    }

    /**
     * Returns the value of epsilon
     * @return double Epsilon
     */
    public double getEpsilon() {
        return epsilon;
    }

    /**
     * Returns the value of minPoints
     * @return int MinPoints
     */
    public int getMinPoints() {
        return minPoints;
    }

    /**
     * Returns the distance-type
     * @return String Distance-type
     */
    public String getDatabase_distanceType() {
        return database_distanceType;
    }

    /**
     * Returns the type of the used index (database)
     * @return String Index-type
     */
    public String getDatabase_Type() {
        return database_Type;
    }

    /**
     * Sets a new distance-type
     * @param database_distanceType The new distance-type
     */
    public void setDatabase_distanceType(String database_distanceType) {
        this.database_distanceType = database_distanceType;
    }

    /**
     * Sets a new database-type
     * @param database_Type The new database-type
     */
    public void setDatabase_Type(String database_Type) {
        this.database_Type = database_Type;
    }

    /**
     * Returns the tip text for this property
     * @return tip text for this property suitable for
     * displaying in the explorer/experimenter gui
     */
    public String epsilonTipText() {
        return "radius of the epsilon-range-queries";
    }

    /**
     * Returns the tip text for this property
     * @return tip text for this property suitable for
     * displaying in the explorer/experimenter gui
     */
    public String minPointsTipText() {
        return "minimun number of DataObjects required in an epsilon-range-query";
    }

    /**
     * Returns the tip text for this property
     * @return tip text for this property suitable for
     * displaying in the explorer/experimenter gui
     */
    public String database_TypeTipText() {
        return "used database";
    }

    /**
     * Returns the tip text for this property
     * @return tip text for this property suitable for
     * displaying in the explorer/experimenter gui
     */
    public String database_distanceTypeTipText() {
        return "used distance-type";
    }

    /**
     * Returns a string describing this DataMining-Algorithm
     * @return String Information for the gui-explorer
     */
    public String globalInfo() {
        return getTechnicalInformation().toString();
    }

    /**
     * Returns an instance of a TechnicalInformation object, containing 
     * detailed information about the technical background of this class,
     * e.g., paper reference or book this class is based on.
     * 
     * @return the technical information about this class
     */
    public TechnicalInformation getTechnicalInformation() {
      TechnicalInformation 	result;
      
      result = new TechnicalInformation(Type.INPROCEEDINGS);
      result.setValue(Field.AUTHOR, "Martin Ester and Hans-Peter Kriegel and Joerg Sander and Xiaowei Xu");
      result.setValue(Field.TITLE, "A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise");
      result.setValue(Field.BOOKTITLE, "Second International Conference on Knowledge Discovery and Data Mining");
      result.setValue(Field.EDITOR, "Evangelos Simoudis and Jiawei Han and Usama M. Fayyad");
      result.setValue(Field.YEAR, "1996");
      result.setValue(Field.PAGES, "226-231");
      result.setValue(Field.PUBLISHER, "AAAI Press");
      
      return result;
    }

    /**
     * Returns a description of the clusterer
     * 
     * @return a string representation of the clusterer
     */
    public String toString() {
        StringBuffer stringBuffer = new StringBuffer();
        stringBuffer.append("DBScan clustering results\n" +
                "========================================================================================\n\n");
        stringBuffer.append("Clustered DataObjects: " + database.size() + "\n");
        stringBuffer.append("Number of attributes: " + database.getInstances().numAttributes() + "\n");
        stringBuffer.append("Epsilon: " + getEpsilon() + "; minPoints: " + getMinPoints() + "\n");
        stringBuffer.append("Index: " + getDatabase_Type() + "\n");
        stringBuffer.append("Distance-type: " + getDatabase_distanceType() + "\n");
        stringBuffer.append("Number of generated clusters: " + numberOfGeneratedClusters + "\n");
        DecimalFormat decimalFormat = new DecimalFormat(".##");
        stringBuffer.append("Elapsed time: " + decimalFormat.format(elapsedTime) + "\n\n");

        for (int i = 0; i < database.size(); i++) {
            DataObject dataObject = database.getDataObject(Integer.toString(i));
            stringBuffer.append("(" + Utils.doubleToString(Double.parseDouble(dataObject.getKey()),
                    (Integer.toString(database.size()).length()), 0) + ".) "
                    + Utils.padRight(dataObject.toString(), 69) + "  -->  " +
                    ((dataObject.getClusterLabel() == DataObject.NOISE) ?
                    "NOISE\n" : dataObject.getClusterLabel() + "\n"));
        }
        return stringBuffer.toString() + "\n";
    }

    /**
     * Main Method for testing DBScan
     * @param args Valid parameters are: 'E' epsilon (default = 0.9); 'M' minPoints (default = 6);
     *                                   'I' index-type (default = weka.clusterers.forOPTICSAndDBScan.Databases.SequentialDatabase);
     *                                   'D' distance-type (default = weka.clusterers.forOPTICSAndDBScan.DataObjects.EuclidianDataObject);
     */
    public static void main(String[] args) {
        runClusterer(new DBScan(), args);
    }

    // *****************************************************************************************************************
    // inner classes
    // *****************************************************************************************************************

}