本文介绍如何在Spark上使用Scala实现SimRank算法。通过读取数据，构建图结构，迭代计算相似度，并最终保存结果，展示了大数据环境下SimRank的并行计算过程。
package main.scala
import org.apache.spark._
import SparkContext._
object SparkSimRank {
def main(args: Array[String]) {
val iters = 10
val conf = new SparkConf()
.setMaster("local")
.setAppName("PagRank")
.set("spark.executor.memory", "1g")
.setSparkHome("/home/wang/spark-0.9.0") //(System.getenv("SPARK_HOME"))
.setJars(Seq(System.getenv("SPARK_TEST_JAR")))
val cs = new SparkContext(conf)
val simrankline = cs.textFile("/home/wang/Desktop/experiment/input/5")
val graphline = cs.textFile("/home/wang/Desktop/input/5")
val indegreeline = cs.textFile("/home/wang/Desktop//input1/5")
val graph = graphline.map{ s =>
val key = s.spli
import org.apache.spark._
import SparkContext._
object SparkSimRank {
def main(args: Array[String]) {
val iters = 10
val conf = new SparkConf()
.setMaster("local")
.setAppName("PagRank")
.set("spark.executor.memory", "1g")
.setSparkHome("/home/wang/spark-0.9.0") //(System.getenv("SPARK_HOME"))
.setJars(Seq(System.getenv("SPARK_TEST_JAR")))
val cs = new SparkContext(conf)
val simrankline = cs.textFile("/home/wang/Desktop/experiment/input/5")
val graphline = cs.textFile("/home/wang/Desktop/input/5")
val indegreeline = cs.textFile("/home/wang/Desktop//input1/5")
val graph = graphline.map{ s =>
val key = s.spli
最低0.47元/天 解锁文章
1万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
