基于大数据spark+hadoop的图书可视化分析系统

作者主页：计算机毕设老哥
精彩专栏推荐订阅：在 下方专栏

Java实战项目专栏

Python实战项目专栏

安卓实战项目专栏

微信小程序实战项目专栏

一、开发介绍

1.1 开发环境

• 技术栈：spark+hadoop+hive

• 离线ETL+在线数据分析 (OLAP)+流计算+机器学习+图计算

二、系统介绍

2.1图片展示

注册登录页面：大数据可视化分析模块：管理员后面模块：

三、部分代码设计

package org.pact518

import org.apache.spark.{SparkConf, SparkContext}
import org.apache.spark.mllib.recommendation._
import org.apache.spark.rdd.RDD
import scala.math.sqrt
import org.jblas.DoubleMatrix

object alsBatchRecommender {
  private val minSimilarity = 0.6

  def cosineSimilarity(vector1: DoubleMatrix, vector2: DoubleMatrix): Double = vector1.dot(vector2) / (vector1.norm2() * vector2.norm2())

  def calculateAllCosineSimilarity(model: MatrixFactorizationModel, dataDir: String, dateStr: String): Unit = {
    //calculate all the similarity and store the stuff whose sim > 0.5 to Redis.
    val productsVectorRdd = model.productFeatures
      .map{case (movieId, factor) =>
      val factorVector = new DoubleMatrix(factor)
      (movieId, factorVector)
    }
    
    val productsSimilarity = productsVectorRdd.cartesian(productsVectorRdd)
      .filter{ case ((movieId1, vector1), (movieId2, vector2)) => movieId1 != movieId2 }
      .map{case ((movieId1, vector1), (movieId2, vector2)) =>
        val sim = cosineSimilarity(vector1, vector2)
        (movieId1, movieId2, sim)
      }.filter(_._3 >= minSimilarity)
    
    productsSimilarity.map{ case (movieId1, movieId2, sim) => 
      movieId1.toString + "," + movieId2.toString + "," + sim.toString
    }.saveAsTextFile(dataDir + "allSimilarity_" + dateStr)

    productsVectorRdd.unpersist()
    productsSimilarity.unpersist()
  }

  def main(args: Array[String]) {
    val conf = new SparkConf().setAppName("alsBatchRecommender").set("spark.executor.memory", "2g")
    val sc = new SparkContext(conf)
    if (args.length < 1) {
        println("USAGE:")
        println("spark-submit ... xxx.jar Date_String [Iteration]")
        println("spark-submit ... xxx.jar 20160424 10")
        sys.exit()
    }
    val dateStr = args(0)

    val iterations = if (args.length > 1) args(1).toInt else 5

    val dataDir = "hdfs://master:9001/leechanx/netflix/"

    val trainData = sc.textFile(dataDir + "trainingData.txt").map{ line =>
      val lineAttrs = line.trim.split(",")
      Rating(lineAttrs(1).toInt, lineAttrs(0).toInt, lineAttrs(2).toDouble)
    }.cache()

    val (rank, lambda) = (50, 0.01)
    val model = ALS.train(trainData, rank, iterations, lambda)

    trainData.unpersist()

    calculateAllCosineSimilarity(model, dataDir, dateStr) //save cos sim.
    model.save(sc, dataDir + "ALSmodel_" + dateStr) //save model.

    val realRatings = sc.textFile(dataDir + "realRatings.txt").map{ line =>
      val lineAttrs = line.trim.split(",")
      Rating(lineAttrs(1).toInt, lineAttrs(0).toInt, lineAttrs(2).toDouble)
    }

    val rmse = computeRmse(model, realRatings)
    println("the Rmse = " + rmse)

    sc.stop()
  }

  def parameterAdjust(trainData: RDD[Rating], realRatings: RDD[Rating]): (Int, Double, Double) = {
    val evaluations =
      for (rank   <- Array(10,  50);
           lambda <- Array(1.0, 0.0001);
           alpha  <- Array(1.0, 40.0))
        yield {
          val model = ALS.trainImplicit(trainData, rank, 10, lambda, alpha)
          val rmse = computeRmse(model, realRatings)
          unpersist(model)
          ((rank, lambda, alpha), rmse)
        }
    val ((rank, lambda, alpha), rmse) = evaluations.sortBy(_._2).head
    println("After parameter adjust, the best rmse = " + rmse)
    (rank, lambda, alpha)
  }

  def computeRmse(model: MatrixFactorizationModel, realRatings: RDD[Rating]): Double = {
    val testingData = realRatings.map{ case Rating(user, product, rate) =>
      (user, product)
    }

    val prediction = model.predict(testingData).map{ case Rating(user, product, rate) =>
      ((user, product), rate)
    }

    val realPredict = realRatings.map{case Rating(user, product, rate) =>
      ((user, product), rate)
    }.join(prediction)

    sqrt(realPredict.map{ case ((user, product), (rate1, rate2)) =>
      val err = rate1 - rate2
      err * err
    }.mean())//mean = sum(list) / len(list)
  }

  def unpersist(model: MatrixFactorizationModel): Unit = {
    // At the moment, it's necessary to manually unpersist the RDDs inside the model
    // when done with it in order to make sure they are promptly uncached
    model.userFeatures.unpersist()
    model.productFeatures.unpersist()
  }
}

<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>
    <groupId>com.github.hellowzk</groupId>
    <artifactId>light-spark</artifactId>
    <version>1.0.4-release</version>
    <packaging>pom</packaging>
    <!--<parent>
        <groupId>org.sonatype.oss</groupId>
        <artifactId>oss-parent</artifactId>
        <version>7</version>
    </parent>-->
    <modules>

        <module>light-spark-core</module>
        <module>assembly</module>
        <module>example</module>
    </modules>

    <properties>
        <project.reporting.outputEncoding>UTF-8</project.reporting.outputEncoding>
        <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
        <java.encoding>${project.build.sourceEncoding}</java.encoding>
        <java.version>1.8</java.version>
        <scala.encoding>${project.build.sourceEncoding}</scala.encoding>
        <scala.version.major>2.11</scala.version.major>
        <scala.version>${scala.version.major}.12</scala.version>
        <spark.version.pre>2.2</spark.version.pre>
        <spark.version>${spark.version.pre}.3</spark.version>
        <hbase.version>1.4.3</hbase.version>
        <hadoop.version>2.6.0-cdh5.13.0</hadoop.version>
        <spark.deploy.mode>cluster</spark.deploy.mode>
        <app.build.profile.id></app.build.profile.id>
    </properties>

    <build>
        <plugins>
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-resources-plugin</artifactId>
                <version>3.0.2</version>
                <configuration>
                    <nonFilteredFileExtensions>
                        <nonFilteredFileExtension>sh</nonFilteredFileExtension>
                        <nonFilteredFileExtension>conf</nonFilteredFileExtension>
                        <nonFilteredFileExtension>json</nonFilteredFileExtension>
                        <nonFilteredFileExtension>txt</nonFilteredFileExtension>
                        <nonFilteredFileExtension>xlsx</nonFilteredFileExtension>
                    </nonFilteredFileExtensions>
                </configuration>
            </plugin>

            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-compiler-plugin</artifactId>
                <version>3.7.0</version>
                <configuration>
                    <source>${java.version}</source>
                    <target>${java.version}</target>
                </configuration>
            </plugin>

            <plugin>
                <groupId>net.alchim31.maven</groupId>
                <artifactId>scala-maven-plugin</artifactId>
                <version>3.2.1</version>
                <executions>
                    <execution>
                        <id>scala-compile-first</id>
                        <phase>process-resources</phase>
                        <goals>
                            <goal>add-source</goal>
                            <goal>compile</goal>
                        </goals>
                    </execution>
                    <execution>
                        <id>scala-test-compile</id>
                        <phase>process-test-resources</phase>
                        <goals>
                            <goal>testCompile</goal>
                        </goals>
                    </execution>
                </executions>
                <configuration>
                    <jvmArgs>
                        <jvmArg>-Xms64m</jvmArg>
                        <!--可以在测试环境修改为1024m,提交时为2048m-->
                        <!--<jvmArg>-Xmx1024m</jvmArg>-->
                        <jvmArg>-Xmx2048m</jvmArg>
                    </jvmArgs>
                    <scalaVersion>${scala.version}</scalaVersion>
                    <args>
                        <arg>-target:jvm-${java.version}</arg>
                    </args>
                </configuration>
            </plugin>

            <!--生成javadoc包的插件-->
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-javadoc-plugin</artifactId>
                <version>2.9.1</version>
                <executions>
                    <execution>
                        <phase>package</phase>
                        <goals>
                            <goal>jar</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
            <!--生成java源码包插件（仅对java有用，对scala不管用） source-->
            <plugin>
                <groupId>org.apache.maven.plugins</groupId>
                <artifactId>maven-source-plugin</artifactId>
                <version>3.0.1</version>
                <executions>
                    <execution>
                        <id>attach-sources</id>
                        <goals>
                            <goal>jar-no-fork</goal>
                        </goals>
                    </execution>
                </executions>
            </plugin>
        </plugins>

    </build>

    <profiles>
        <profile>
            <id>dev</id>
            <properties>
                <app.build.profile.id>dev</app.build.profile.id>
            </properties>
        </profile>
        <profile>
            <id>spark2.2</id>
            <activation>
                <activeByDefault>true</activeByDefault>
            </activation>
            <properties>
                <spark.version.pre>2.2</spark.version.pre>
            </properties>
        </profile>
        <profile>
            <id>spark2.3</id>
            <properties>
                <spark.version.pre>2.3</spark.version.pre>
            </properties>
        </profile>
        <profile>
            <id>spark2.4</id>
            <properties>
                <spark.version.pre>2.4</spark.version.pre>
            </properties>
        </profile>
    </profiles>

    <distributionManagement>
        <snapshotRepository>
            <id>ossrh</id>
            <url>https://oss.sonatype.org/content/repositories/snapshots/</url>
        </snapshotRepository>
        <repository>
            <id>ossrh</id>
            <url>https://oss.sonatype.org/service/local/staging/deploy/maven2/</url>
        </repository>
    </distributionManagement>
</project>

总结

大家可以帮忙点赞、收藏、关注、评论啦

有问题评论区交流

精彩专栏推荐订阅：在 下方专栏

Java实战项目专栏

Python实战项目专栏

安卓实战项目专栏

微信小程序实战项目专栏