/**
* 执行KNN算法，获取测试元组的类别
*
* @param list 训练数据集
* @param test 测试元组
* @param k 设定的K值
* @return 测试元组的类别
*/
public String knn(List<List<Double>> list, List<Double> test, int k) {
PriorityQueue<KNNNode> pq = new PriorityQueue<KNNNode>(k, comparator);
for (int i = 0; i < k; i++) {
List<Double> trainTuple = list.get(i);
Double c = trainTuple.get(0);
KNNNode node = new KNNNode(i, getDistance(test, trainTuple), c);
pq.add(node);
}
for (int i = 0; i < list.size(); i++) {
List<Double> t = list.get(i);
double distance = getDistance(test, t);
KNNNode top = pq.peek();
if (top.getDistance() > distance) {
pq.remove();
pq.add(new KNNNode(i, distance, t.get(0)));
}
}
return getMostClass(pq);
}
/**
* 获取所得到的k个最近邻元组的多数类
*
* @param pq 存储k个最近近邻元组的优先级队列
* @return 多数类的名称
*/
private String getMostClass(PriorityQueue<KNNNode> pq) {
Map<Double, Integer> classCount = new HashMap<Double, Integer>();
for (int i = 0; i < pq.size(); i++) {
KNNNode node = pq.remove();
double c = node.getC();
if (classCount.containsKey(c)) {
classCount.put(c, classCount.get(c) + 1);
} else {
classCount.put(c, 1);
}
}
int maxIndex = -1;
int maxCount = 0;
Object[] classes = classCount.keySet().toArray();
for (int i = 0; i < classes.length; i++) {
if (classCount.get(classes[i]) > maxCount) {
maxIndex = i;
maxCount = classCount.get(classes[i]);
}
}
return classes[maxIndex].toString();
}