/**
* Tests if the database contains the dataObject_Query
*
* @param dataObject_Query The query-object
* @return true if the database contains dataObject_Query, else false
*/
public boolean contains(DataObject dataObject_Query) {
Iterator iterator = dataObjectIterator();
while (iterator.hasNext()) {
DataObject dataObject = (DataObject) iterator.next();
if (dataObject.equals(dataObject_Query)) return true;
}
return false;
}
/**
* Emits the k next-neighbours and performs an epsilon-range-query at the parallel. The returned
* list contains two elements: At index=0 --> list with all k next-neighbours; At index=1 --> list
* with all dataObjects within epsilon;
*
* @param k number of next neighbours
* @param epsilon Specifies the range for the query
* @param dataObject the start object
* @return list with the k-next neighbours (PriorityQueueElements) and a list with candidates from
* the epsilon-range-query (EpsilonRange_ListElements)
*/
public List k_nextNeighbourQuery(int k, double epsilon, DataObject dataObject) {
Iterator iterator = dataObjectIterator();
List return_List = new ArrayList();
List nextNeighbours_List = new ArrayList();
List epsilonRange_List = new ArrayList();
PriorityQueue priorityQueue = new PriorityQueue();
while (iterator.hasNext()) {
DataObject next_dataObject = (DataObject) iterator.next();
double dist = dataObject.distance(next_dataObject);
if (dist <= epsilon)
epsilonRange_List.add(new EpsilonRange_ListElement(dist, next_dataObject));
if (priorityQueue.size() < k) {
priorityQueue.add(dist, next_dataObject);
} else {
if (dist < priorityQueue.getPriority(0)) {
priorityQueue.next(); // removes the highest distance
priorityQueue.add(dist, next_dataObject);
}
}
}
while (priorityQueue.hasNext()) {
nextNeighbours_List.add(0, priorityQueue.next());
}
return_List.add(nextNeighbours_List);
return_List.add(epsilonRange_List);
return return_List;
}
/**
* Sets the minimum and maximum values for each attribute in different arrays by walking through
* every DataObject of the database
*/
public void setMinMaxValues() {
attributeMinValues = new double[getInstances().numAttributes()];
attributeMaxValues = new double[getInstances().numAttributes()];
// Init
for (int i = 0; i < getInstances().numAttributes(); i++) {
attributeMinValues[i] = attributeMaxValues[i] = Double.NaN;
}
Iterator iterator = dataObjectIterator();
while (iterator.hasNext()) {
DataObject dataObject = (DataObject) iterator.next();
for (int j = 0; j < getInstances().numAttributes(); j++) {
if (Double.isNaN(attributeMinValues[j])) {
attributeMinValues[j] = dataObject.getInstance().value(j);
attributeMaxValues[j] = dataObject.getInstance().value(j);
} else {
if (dataObject.getInstance().value(j) < attributeMinValues[j])
attributeMinValues[j] = dataObject.getInstance().value(j);
if (dataObject.getInstance().value(j) > attributeMaxValues[j])
attributeMaxValues[j] = dataObject.getInstance().value(j);
}
}
}
}
/**
* Performs an epsilon range query for this dataObject
*
* @param epsilon Specifies the range for the query
* @param queryDataObject The dataObject that is used as query-object for epsilon range query
* @return List with all the DataObjects that are within the specified range
*/
public List epsilonRangeQuery(double epsilon, DataObject queryDataObject) {
ArrayList epsilonRange_List = new ArrayList();
Iterator iterator = dataObjectIterator();
while (iterator.hasNext()) {
DataObject dataObject = (DataObject) iterator.next();
double distance = queryDataObject.distance(dataObject);
if (distance < epsilon) {
epsilonRange_List.add(dataObject);
}
}
return epsilonRange_List;
}
/**
* @param args
* @throws Exception
*/
void analyseResult(DBScan algo, Instances data) throws Exception {
FileWriter fw = new FileWriter(new File("topictrackresult"));
BufferedWriter bw = new BufferedWriter(fw);
StringBuilder sb = new StringBuilder();
FastVector resultset = new FastVector();
for (int i = 0; i < algo.numberOfClusters(); i++) {
ArrayList<String> oneCluster = new ArrayList<String>();
resultset.addElement(oneCluster);
}
for (int i = 0; i < algo.database.size(); i++) {
DataObject dataObject = algo.database.getDataObject(Integer.toString(i));
int IDraw = Integer.parseInt(dataObject.getKey());
int ID = (int) data.instance(IDraw).value(0);
String title = findFileNameWithID(ID);
if (DataObject.NOISE != dataObject.getClusterLabel()) {
int label = dataObject.getClusterLabel();
ArrayList<String> filenamelib = (ArrayList<String>) resultset.elementAt(label);
filenamelib.add(title);
}
}
sb.append("The generated Clusters are:\n");
for (int i = 0; i < algo.numberOfClusters(); i++) {
sb.append("\r\n cluster" + i + "\n");
ArrayList<String> filenamelib = (ArrayList<String>) resultset.elementAt(i);
Iterator iter = filenamelib.iterator();
while (iter.hasNext()) {
String title = (String) iter.next();
sb.append(title + "\n");
}
}
bw.write(sb.toString());
bw.close();
fw.close();
}
/**
* Inserts a new dataObject into the database
*
* @param dataObject
*/
public void insert(DataObject dataObject) {
treeMap.put(dataObject.getKey(), dataObject);
}