【机器学习笔记五】聚类 - k均值聚类

package kmean

import org.apache.spark.mllib.clustering.KMeans
import org.apache.spark.mllib.linalg.Vectors
import org.apache.spark.{SparkContext, SparkConf}

/*-
 * kmean聚类
 * Fredric 2017
 */
object kmean {
  def main(args:Array[String]): Unit ={

    val conf = new SparkConf().setMaster("local").setAppName("kmean")
    val sc   = new SparkContext(conf)

    //每一个向量代表特征空间中的一个点
    val v0  = Vectors.dense(1)
    val v1  = Vectors.dense(2)
    val v2  = Vectors.dense(3)
    val v3  = Vectors.dense(4)

    val v4  = Vectors.dense(11)
    val v5  = Vectors.dense(25)
    val v6  = Vectors.dense(31)

    val data = sc.parallelize(Seq(v0, v1, v2, v3, v4, v5, v6))

    //setK表示最后聚类的分组数量是2，分成2组
    val model = new KMeans().setMaxIterations(20).setK(2).run(data)

/*
    输出的两个中心点如下：
    [4.2]
    [28.0]*/
    model.clusterCenters.foreach(println)
  }
}