javaants.java~

蚁群聚类的java源码ant colony clustering by java

		} catch (Exception e) {
			System.err.println("Exception in Main Routine");
			System.err.println(e.getMessage());
			System.exit(0);
		}
	}
 

/**************** Equivalent initialisation once some form of data is provided ********************************/

	private static void init(boolean old, String filename) {
		DistanceMatrix d = null;

		keywords = new String[data.getnkeys()];
		documents = new Document[data.getndocs()];
		System.out.println("Keywords: "+ data.getnkeys());
		System.out.println("Documents: "+ data.getndocs());
		
		// store data	
		keywords = data.getKeys();
		documents = data.getDocs();
		configuration.setndocs(data.getndocs());
		configuration.setnkeys(data.getnkeys());
		
		
			
		// remove old maps
		if (map != null) frame.remove(map);
			



		// create the new topic map			
		map = new TopicMap(configuration, keywords, documents, antMode, d);
	
		
		// update frame
		frame.add(map, BorderLayout.CENTER);
		frame.pack();
   	 	frame.setSize(configuration.getxsize()*5+10,configuration.getysize()*5+48);
   	 	frame.show();
		frame.setVisible(true);
	   	init = true;
		if ((antMode == false) || (old == true)) ;
		else new Thread(map).start();
	}
		

 
/************** I/O Functions **********************************************************************/

	/** Save a document collection to disk
	* @param data the provided document collections
	* @param filename the name of the file to write to
	*/
	public static void save(Data data, String filename) {
		try {
			ObjectOutputStream output = new ObjectOutputStream(new FileOutputStream(filename));
			output.writeInt(data.getndocs());
			output.writeInt(data.getnkeys());
			output.writeObject(data.getDocs());
			output.writeObject(data.getKeys());

			output.flush();
			output.close();
		} catch (IOException e) {
			System.out.println("Error occured when writing from file " + filename);
			System.out.println(e.getMessage());
		}
		return;
	}

	/** Function to read a document collection from disk
	* @param filename the name of the file to read from
	* @return the loaded document collection
	*/
	public static Data read(String filename) {
		Data data = null;
		Document [] docs;
		String [] keys;
		int ndocs, nkeys;
		try {
			ObjectInputStream input = new ObjectInputStream(new FileInputStream(filename));
			ndocs = input.readInt();
			nkeys = input.readInt();
			docs = (Document [])input.readObject();
			keys = (String [])input.readObject();
			input.close();
			data = new Data(nkeys, ndocs, keys, docs);
		} catch (Exception e) {
			System.out.println("Error occured when reading from file " + filename);
			System.out.println(e.getMessage());
		}
		return data;
	}



/*********** hardcoded generation of artificial data *************************************/

/** Generate the hardcoded data for the test distributions
* @param dist switch between the three possible distributions (0, 1 or 2)
* @return the generated test data
*/
public static Data generate(int dist) {
	int ndocs = 400;
	int nkeys = 2;

	String [] keys = new String[2];
	keys[0] = new String("ycoord");
	keys[1] = new String("xcoord");
		
	Document[] docs = null;

	


	// Hard-coded test distribution (Uniform Distribution)
	if (dist == 1) {

		ndocs = 450;
		docs = new Document[ndocs + nkeys];
		Random generator = new Random();

		Position center[] = new Position[9];
		center[0] = new Position(0, 0);
		center[1] = new Position(0, 8);
		center[2] = new Position(8, 0);
		center[3] = new Position(8, 8);
		center[4] = new Position(0, 16);
		center[5] = new Position(16, 0);
		center[6] = new Position(16, 8);
		center[7] = new Position(8, 16);
		center[8] = new Position(16, 16);
		Position var[] = new Position [10];
		var[0] = new Position(2, 2);
		var[1] = new Position(2, 2);
		var[2] = new Position(2, 2);
		var[3] = new Position(2, 2);
		var[4] = new Position(2, 2);
		var[5] = new Position(2, 2);
		var[6] = new Position(2, 2);
		var[7] = new Position(2, 2);
		var[8] = new Position(2, 2);
				
		double data[] = new double[2];
		Position pos;
		
		// create individual vectors distributed randomly around the given mean
		for (int j=0; j< 50; j++) {
			for (int i = 0; i< 9; i++) {
				data[0] = Math.random() * 4*var[i].getY() + center[i].getY();
				data[1] = Math.random() * 4*var[i].getX() + center[i].getX();
				pos = new Position((int)Math.round(data[0]), (int)Math.round(data[1]));
				docs[j*9+i] = new Document(data, 2, pos, i+1);
			}
		}
		

		
	}
	
	// Hard-coded test distribution (Normal Distribution)
	else if (dist == 0) {
		
	        ndocs = 800;
		nkeys = 2;

		docs = new Document[ndocs + nkeys];
		Random generator = new Random();

		Position center[] = new Position[4];
		center[0] = new Position(0, 0);
		center[1] = new Position(0, 8);
		center[2] = new Position(8, 0);
		center[3] = new Position(8, 8);
		Position var[] = new Position [4];
		var[0] = new Position(2, 2);
		var[1] = new Position(2, 2);
		var[2] = new Position(2, 2);
		var[3] = new Position(2, 2);
	
		double data[] = new double[2];
		Position pos;
		
		for (int i = 0; i< 4; i++) {
			for (int j=0; j< 200 ; j++) {
				data[0] = generator.nextGaussian() * var[i].getY() + center[i].getY();
				data[1] = generator.nextGaussian() * var[i].getX() + center[i].getX();
				pos = new Position((int)Math.round(data[0]), (int)Math.round(data[1]));
				docs[i*200+j] = new Document(data, 2, pos, i+1);

			}
		}



	}

	else if (dist == 2) {

		ndocs = 90;
		nkeys = 20;
		docs = new Document[ndocs+nkeys];
		
		
		keys = new String[nkeys];
		keys[0] = "Computer Graphics";
		keys[1] = "Splines";
		keys[2] = "Mesh";
		keys[3] = "Visualization";
		keys[4] = "Topic Maps";
		keys[5] = "Fisheye Views";
		keys[6] = "Graphs";

		keys[7] = "Artificial Life";
		keys[8] = "Game of Life";
		keys[9] = "Turing";
		keys[10] = "Chaos";
		keys[11] = "Self-organisation";

		keys[12] = "Artificial Intelligence";
		keys[13] = "Bayesian Reasoning";
		keys[14] = "Symbolic";
		keys[15] = "Expert System";
		keys[16] = "Neural Network";
		keys[17] = "Knowledge";
		keys[18] = "Representation";
		keys[19] = "Robot";


		double data[] = new double[nkeys];
			
		for (int j=0; j<ndocs; j++) {
		
			int color = 1;
			if (j < 30) {
				for (int k=0; k<7; k++) {
					data[k] =  (int)Math.floor(3.0+Math.random()*3.0);
				}
				for (int k=7; k<20; k++) {
					data[k] = (int)Math.floor(Math.random()*3.0);
				}
				color = 2;
			}
			else if (j < 60) {
				for (int k=0; k<7; k++) {
					data[k] = (int)Math.floor(Math.random()*3.0);
				}
				for (int k=7; k<12; k++) {
					data[k] = (int)Math.floor(3.0+Math.random()*10.0);
				}
				for (int k=12; k<20; k++) {
					data[k] = (int)Math.floor(Math.random()*3.0);
				}
				color = 3;
			}	
			else {
				for (int k=0; k<12; k++) {
					data[k] = (int)Math.floor(Math.random()*3.0);
				}
				for (int k=12; k<20; k++) {
					data[k] = (int)Math.floor(30.0+Math.random()*5.0);
				}
				color = 4;
			}
		
			docs[j] = new Document(color);
			docs[j].setVector(nkeys, data);
		}


			
		
	}
	return new Data(nkeys, ndocs, keys, docs);
	}

}