package cn.edu.bjut.chapter4;

import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Random;

import org.apache.commons.math3.ml.clustering.Cluster;
import org.apache.commons.math3.ml.clustering.DBSCANClusterer;

public class DBSCANClustering {
	public static List<Cluster<DocumentPoint>> cluster(double epsilon, int minPoints, List<String> documents) {
		// Suppose we have the points ready
		List<DocumentPoint> points = new ArrayList<DocumentPoint>();
		Random rand = new Random(0);
		for (String document : documents) {
			// Here we could represent document as points based on TF-IDF
			// For simplicity, we will use random values or replace with actual TF-IDF
			// vectors.
			double[] randomPoints = new double[] { rand.nextDouble(), rand.nextDouble() };
			points.add(new DocumentPoint(randomPoints));
		}

		// Populate DocumentPoints from documents as before
		DBSCANClusterer<DocumentPoint> dbscan = new DBSCANClusterer<DocumentPoint>(epsilon, minPoints);
		return dbscan.cluster(points);
	}

	public static void main(String[] args) {
		int k = 5;
		String[] documents = { "Java is a high-level, class-based, object-oriented programming language.",
				"Python is an interpreted, high-level and general-purpose programming language.",
				"JavaScript is a programming language that conforms to the ECMAScript specification.",
				"C++ is a general-purpose programming language created by Bjarne Stroustrup.",
				"PHP is a popular general-purpose scripting language that is especially suited to web development.",
				"Ruby is a dynamic, open source programming language with a focus on simplicity and productivity." };
		List<String> documentList = Arrays.asList(documents);

		List<Cluster<DocumentPoint>> clusters = DBSCANClustering.cluster(0.5, 1, documentList);
		for (Cluster<DocumentPoint> cluster : clusters) {
			for (DocumentPoint doc : cluster.getPoints()) {
				System.out.println(doc);
			}

			System.out.println();
		}
	}
}