基于Spark的电信客户流失数据分析系统毕设mapreduce_hive
舟率率 4/20/2026 springbootjavasqoopmapreducevue
# 项目概况
mapreduce_hive (opens new window)
# 数据类型
电信客户流失数据
# 开发环境
centos7
# 软件版本
hadoop3.2.0、hive3.1.2、spark3.1.2、mysql5.7.38、scala2.12.18、jdk8、sqoop1.4.7
# 开发语言
Scala、Java
# 开发流程
数据上传(hdfs)->数据清洗(mapreduce)->数据分析(hive)->模型训练集预测(spark)->数据存储(mysql)->后端(springboot)->前端(vue)
# 可视化图表
# 操作步骤
# 启动MySQL
# 查看mysql是否启动 启动命令: systemctl start mysqld.service
systemctl status mysqld.service
# 进入mysql终端
# MySQL的用户名:root 密码:123456
# MySQL的用户名:root 密码:123456
# MySQL的用户名:root 密码:123456
mysql -uroot -p123456
1
2
3
4
5
6
7
8
9
2
3
4
5
6
7
8
9
# 启动Hadoop
# 离开安全模式: hdfs dfsadmin -safemode leave
# 启动hadoop
bash /export/software/hadoop-3.2.0/sbin/start-hadoop.sh
1
2
3
4
5
2
3
4
5
# 启动hive
# 在第一个窗口中,执行后等待10-20秒
/export/software/apache-hive-3.1.2-bin/bin/hive --service metastore
# 在第二个窗口中,执行后等待10-20秒
/export/software/apache-hive-3.1.2-bin/bin/hive --service hiveserver2
# 连接进入hive终端命令如下:
# /export/software/apache-hive-3.1.2-bin/bin/beeline -u jdbc:hive2://master:10000 -n root
1
2
3
4
5
6
7
8
9
10
2
3
4
5
6
7
8
9
10
# 准备目录
mkdir -p /data/jobs/project/
cd /data/jobs/project/
# 解压 "project-spark-telecom-customer-churn-data-analysis-system/" 目录下的 "data.7z" 到当前目录下
# 上传 "project-spark-telecom-customer-churn-data-analysis-system" 整个文件夹 到 "/data/jobs/project/" 目录
1
2
3
4
5
6
7
2
3
4
5
6
7
# 初始化MySQL表
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/mysql/
mysql -uroot -p123456 < mysql.sql
1
2
3
4
5
2
3
4
5
# 上传文件到hdfs
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/
hdfs dfs -mkdir -p /data/input/
hdfs dfs -rm -r /data/input/*
hdfs dfs -put telecom_churn_large.csv /data/input/
hdfs dfs -ls /data/input/
1
2
3
4
5
6
7
8
2
3
4
5
6
7
8
# 程序打包
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
# 对 "spark-job" 目录下的项目 "spark-job" 进行打包
# 打包命令
mvn clean package -DskipTests
1
2
3
4
5
6
7
2
3
4
5
6
7
# 数据清洗
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
spark-submit --master local[2] --class org.example.DataCleaner target/spark-job.jar /data/input/ /data/output/clean/
hdfs dfs -ls /data/output/clean/
1
2
3
4
5
6
7
2
3
4
5
6
7
# 数据分析
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
spark-submit --master local[2] --class org.example.DataAnalyzer target/spark-job.jar /data/output/clean/
1
2
3
4
5
2
3
4
5
# 模型训练
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
spark-submit \
--master local[2] \
--class org.example.ChurnModelTrain \
--driver-memory 2g \
--conf spark.executor.memory=2g \
--conf spark.executor.cores=2 \
--conf spark.sql.shuffle.partitions=20 \
--conf spark.default.parallelism=20 \
--conf spark.memory.fraction=0.4 \
--conf spark.memory.storageFraction=0.3 \
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer \
--conf spark.kryoserializer.buffer.max=256m \
--conf spark.sql.adaptive.enabled=true \
--conf spark.sql.adaptive.coalescePartitions.enabled=true \
--conf spark.network.timeout=600s \
--conf spark.executor.heartbeatInterval=60s \
target/spark-job.jar /data/output/clean/ /data/output/model/
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
# 全量预测
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
spark-submit --master local[2] --class org.example.ChurnPredict target/spark-job.jar /data/output/clean/ /data/output/model/
1
2
3
4
5
2
3
4
5
# 预测结果分析
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/spark_ml/spark-job/
spark-submit --master local[2] --class org.example.PredictionAnalyzer target/spark-job.jar
1
2
3
4
5
2
3
4
5
# 启动后端
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/后端/springboot-demo/
# 对 "springboot-demo" 目录下的项目 "springboot-demo" 进行打包
# 打包命令
mvn clean package -DskipTests
java -jar target/springboot-demo-1.0-SNAPSHOT.jar --model.path=/data/output/model --app.fileBaseDir=/data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system
1
2
3
4
5
6
7
8
9
2
3
4
5
6
7
8
9
# 启动前端
cd /data/jobs/project/project-spark-telecom-customer-churn-data-analysis-system/前端/front_ui/
# 安装node
npm install --registry=https://registry.npmmirror.com
chmod -R 755 node_modules/.bin
npm run dev
# http://master:8080/login
# 登录用户: admin
# 登录密码: 123456
1
2
3
4
5
6
7
8
9
10
11
2
3
4
5
6
7
8
9
10
11