基于Hadoop+Hive+SpringBoot+Vue的肥胖风险数据分析系统

本文介绍了一个基于大数据技术栈的数据分析系统，实现了从数据采集、清洗、存储到可视化展示的完整流程。

一、项目概述

1.1 项目背景

肥胖已成为全球性的健康问题。本项目基于肥胖风险数据集，构建了一个数据分析系统，通过多维度数据分析，为健康管理和政策制定提供数据支持。

1.2 技术栈

大数据处理层：

• Hadoop 3.x：分布式存储和计算框架
• MapReduce：分布式数据处理
• Hive：数据仓库和SQL查询引擎
• Sqoop：数据迁移工具

后端服务层：

• Spring Boot 2.x：应用框架
• MyBatis-Plus：ORM框架
• MySQL：关系型数据库

前端展示层：

• Vue.js 2.x：前端框架
• ECharts：数据可视化库
• Element UI：UI组件库

1.3 项目架构

┌─────────────────────────────────────────────────────────┐
│                    前端展示层 (Vue.js)                   │
│  Dashboard | 趋势分析 | 风险因素 | 人口统计 | 生活方式   │
└─────────────────────────────────────────────────────────┘
                            ↓ HTTP API
┌─────────────────────────────────────────────────────────┐
│                  后端服务层 (Spring Boot)                │
│           ObesityAnalysisController + Service           │
└─────────────────────────────────────────────────────────┘
                            ↓ JDBC
┌─────────────────────────────────────────────────────────┐
│                   数据存储层 (MySQL)                      │
│     ADS应用层数据表 (ads_obesity_trend_analysis等)      │
└─────────────────────────────────────────────────────────┘
                            ↑ Sqoop
┌─────────────────────────────────────────────────────────┐
│                  数据仓库层 (Hive)                        │
│  ODS原始层 | DWD明细层 | DWS汇总层 | ADS应用层         │
└─────────────────────────────────────────────────────────┘
                            ↑ MapReduce
┌─────────────────────────────────────────────────────────┐
│                  数据存储层 (HDFS)                       │
│           /obesity_output (清洗后的数据)                 │
└─────────────────────────────────────────────────────────┘

二、数据仓库设计

2.1 分层架构

项目采用数据仓库分层架构：

ODS层（原始数据层）：

• ods_obesity_raw：存储从MapReduce清洗后的原始数据

DWD层（明细数据层）：

• dwd_gender_info：性别维度表
• dwd_age_group_info：年龄分组维度表
• dwd_obesity_level_info：肥胖水平维度表
• dwd_family_history_info：家族史维度表
• dwd_lifestyle_info：生活方式维度表
• dwd_transport_info：交通方式维度表
• dwd_obesity_risk_fact：肥胖风险事实表

DWS层（汇总数据层）：

• dws_gender_obesity：性别肥胖分析表
• dws_age_group_obesity：年龄组肥胖分析表
• dws_obesity_level_distribution：肥胖水平分布表
• dws_lifestyle_obesity：生活方式肥胖分析表
• dws_family_history_obesity：家族史肥胖分析表
• dws_transport_obesity：交通方式肥胖分析表
• dws_bmi_distribution：BMI分布分析表

ADS层（应用数据层）：

• ads_obesity_trend_analysis：肥胖趋势分析表
• ads_obesity_risk_factors：风险因素分析表
• ads_obesity_demographic_analysis：人口统计学分析表
• ads_obesity_lifestyle_analysis：生活方式分析表
• ads_obesity_comprehensive_analysis：综合分析表

2.2 核心表结构

肥胖趋势分析表（ads_obesity_trend_analysis）：

CREATE TABLE IF NOT EXISTS ads_obesity_trend_analysis (
    age_group STRING COMMENT '年龄分组',
    gender STRING COMMENT '性别',
    total_count BIGINT COMMENT '总人数',
    obesity_count BIGINT COMMENT '肥胖人数',
    obesity_rate DOUBLE COMMENT '肥胖率(%)',
    avg_bmi DOUBLE COMMENT '平均BMI',
    avg_risk_score DOUBLE COMMENT '平均风险评分'
)
COMMENT '肥胖趋势分析表'
PARTITIONED BY (dt STRING COMMENT '日期分区')
ROW FORMAT DELIMITED
FIELDS TERMINATED BY '\001'
STORED AS TEXTFILE;

三、核心功能实现

3.1 MapReduce数据清洗

3.1.1 Driver类设计

ObesityDataDriver.java 是MapReduce作业的驱动类，负责作业的配置和提交。

核心代码：

public class ObesityDataDriver {
    
    public static void main(String[] args) throws Exception {
        
        String inputPath = "data/obesity_level.csv";
        String outputPath = "/obesity_output";

        Configuration conf = new Configuration();
        
        conf.set("HADOOP_USER_NAME", "root");
        conf.set("fs.hdfs.impl", "org.apache.hadoop.hdfs.DistributedFileSystem");
        conf.set("fs.file.impl", "org.apache.hadoop.fs.LocalFileSystem");
        
        UserGroupInformation.setConfiguration(conf);
        UserGroupInformation ugi = UserGroupInformation.createRemoteUser("root");
        UserGroupInformation.setLoginUser(ugi);

        boolean useHDFS = false;
        try {
            conf.set("fs.defaultFS", "hdfs://192.168.199.101:8020");
            FileSystem fs = FileSystem.get(conf);
            fs.close();
            useHDFS = true;
            System.out.println("HDFS连接成功，将使用HDFS输出");
        } catch (Exception e) {
            System.out.println("HDFS连接失败: " + e.getMessage());
            System.out.println("将使用本地文件系统");
            useHDFS = false;
        }
        
        Job job = Job.getInstance(conf, "Obesity Data Cleaner");

        job.setJarByClass(ObesityDataDriver.class);
        job.setMapperClass(ObesityDataCleaner.ObesityCleanerMapper.class);
        job.setReducerClass(ObesityDataCleaner.ObesityCleanerReducer.class);

        job.setMapOutputKeyClass(Text.class);
        job.setMapOutputValueClass(NullWritable.class);

        job.setOutputKeyClass(Text.class);
        job.setOutputValueClass(NullWritable.class);
        
        boolean success = job.waitForCompletion(true);
        System.exit(success ? 0 : 1);
    }
}

实现说明：

1. HDFS连接检测：检测HDFS是否可用，失败时使用本地文件系统
2. 用户权限管理：使用UserGroupInformation进行HDFS用户认证
3. 路径配置：支持命令行参数动态配置输入输出路径

3.1.2 Mapper类设计

ObesityDataCleaner.ObesityCleanerMapper 负责数据的清洗和转换。

核心代码：

public static class ObesityCleanerMapper extends Mapper<Object, Text, Text, NullWritable> {
    
    @Override
    protected void map(Object key, Text value, Context context) 
            throws IOException, InterruptedException {
        
        String line = value.toString().trim();
        
        if (line.isEmpty()) {
            return;
        }
        
        String[] fields = line.split(",");
        
        if (fields.length < 17) {
            context.getCounter("DataQuality", "InvalidFieldCount").increment(1);
            return;
        }
        
        try {
            String cleanedLine = cleanAndValidateRecord(fields, context);
            
            if (cleanedLine != null) {
                context.write(new Text(cleanedLine), NullWritable.get());
                context.getCounter("DataQuality", "ValidRecords").increment(1);
            } else {
                context.getCounter("DataQuality", "FilteredRecords").increment(1);
            }
            
        } catch (Exception e) {
            context.getCounter("DataQuality", "ProcessingErrors").increment(1);
        }
    }
}

数据清洗逻辑：

private String cleanAndValidateRecord(String[] fields, Context context) {
    try {
        String genderVal = normalizeGender(fields[gender]);
        double ageVal = Double.parseDouble(fields[age].trim());
        double heightVal = Double.parseDouble(fields[height].trim());
        double weightVal = Double.parseDouble(fields[weight].trim());
        
        if (heightVal < 1.0 || heightVal > 2.5) {
            context.getCounter("DataQuality", "InvalidHeight").increment(1);
            return null;
        }
        
        if (weightVal < 30 || weightVal > 200) {
            context.getCounter("DataQuality", "InvalidWeight").increment(1);
            return null;
        }
        
        if (ageVal < 14 || ageVal > 80) {
            context.getCounter("DataQuality", "InvalidAge").increment(1);
            return null;
        }
        
        double bmi = calculateBMI(weightVal, heightVal);
        String ageGroup = calculateAgeGroup(ageVal);
        int obesityLevelCode = getObesityLevelCode(obesityLevelVal);
        double riskScore = calculateRiskScore(favcVal, fcvcVal, fafVal, smokeVal, familyHistoryVal);
        
        StringBuilder sb = new StringBuilder();
        sb.append(idVal).append(",");
        sb.append(genderVal).append(",");
        sb.append(df.format(ageVal)).append(",");
        sb.append(df.format(heightVal)).append(",");
        sb.append(df.format(weightVal)).append(",");
        sb.append(familyHistoryVal).append(",");
        sb.append(favcVal).append(",");
        sb.append(df.format(fcvcVal)).append(",");
        sb.append(df.format(ncpVal)).append(",");
        sb.append(caecVal).append(",");
        sb.append(smokeVal).append(",");
        sb.append(df.format(ch2oVal)).append(",");
        sb.append(sccVal).append(",");
        sb.append(df.format(fafVal)).append(",");
        sb.append(df.format(tueVal)).append(",");
        sb.append(calcVal).append(",");
        sb.append(mtransVal).append(",");
        sb.append(obesityLevelVal).append(",");
        sb.append(df.format(bmi)).append(",");
        sb.append(ageGroup).append(",");
        sb.append(obesityLevelCode).append(",");
        sb.append(df.format(riskScore));
        
        return sb.toString();
        
    } catch (NumberFormatException e) {
        context.getCounter("DataQuality", "ParseError").increment(1);
        return null;
    }
}

BMI计算和风险评分：

private double calculateBMI(double weight, double height) {
    return weight / (height * height);
}

private double calculateRiskScore(int favc, double fcvc, double faf, int smoke, int familyHistory) {
    double score = 0;
    score += (favc == 1) ? 20 : 0;
    score += Math.max(0, (3 - fcvc)) * 10;
    score += Math.max(0, (3 - faf)) * 15;
    score += (smoke == 1) ? 10 : 0;
    score += (familyHistory == 1) ? 25 : 0;
    return Math.min(100, Math.max(0, score));
}

实现说明：

1. 数据验证：对身高、体重、年龄等字段进行校验
2. 数据标准化：对性别、肥胖水平、交通方式等字段进行标准化处理
3. 衍生字段计算：计算BMI、年龄分组、肥胖等级编码、风险评分
4. 质量监控：使用Hadoop Counter统计数据质量指标

3.2 Hive ETL流程

3.2.1 性能优化配置

SET hive.exec.parallel=true;
SET hive.exec.parallel.thread.number=16;
SET hive.auto.convert.join=true;
SET hive.map.aggr=true;
SET hive.groupby.skewindata=true;
SET hive.optimize.skewjoin=true;
SET hive.exec.dynamic.partition=true;
SET hive.exec.dynamic.partition.mode=nonstrict;
SET hive.vectorized.execution.enabled=true;
SET hive.vectorized.execution.reduce.enabled=true;

3.2.2 ODS到DWD层转换

加载性别维度表：

INSERT OVERWRITE TABLE dwd_gender_info
SELECT 
    DISTINCT gender
FROM ods_obesity_raw
WHERE gender IS NOT NULL 
  AND gender != '';

加载年龄分组维度表：

INSERT OVERWRITE TABLE dwd_age_group_info
SELECT 
    age_group,
    MIN(age) AS min_age,
    MAX(age) AS max_age,
    CASE age_group
        WHEN 'Adolescent' THEN '青少年(≤18岁)'
        WHEN 'Young_Adult' THEN '青年(19-30岁)'
        WHEN 'Middle_Aged' THEN '中年(31-50岁)'
        WHEN 'Senior' THEN '中老年(51-65岁)'
        WHEN 'Elderly' THEN '老年(>65岁)'
    END AS description
FROM ods_obesity_raw
WHERE age_group IS NOT NULL 
  AND age_group != ''
GROUP BY age_group;

加载肥胖风险事实表：

INSERT OVERWRITE TABLE dwd_obesity_risk_fact PARTITION (dt='2026-02-02')
SELECT 
    id,
    gender,
    age,
    age_group,
    height,
    weight,
    bmi,
    CASE 
        WHEN bmi < 18.5 THEN 'Underweight'
        WHEN bmi < 25 THEN 'Normal'
        WHEN bmi < 30 THEN 'Overweight'
        ELSE 'Obese'
    END AS bmi_group,
    family_history_with_overweight,
    favc,
    fcvc,
    ncp,
    caec,
    smoke,
    ch2o,
    scc,
    faf,
    tue,
    calc,
    mtrans,
    obesity_level,
    obesity_level_code,
    risk_score
FROM ods_obesity_raw;

3.2.3 DWS层汇总计算

性别肥胖分析汇总：

INSERT OVERWRITE TABLE dws_gender_obesity PARTITION (dt='2026-02-02')
SELECT 
    gender,
    COUNT(*) AS total_count,
    SUM(CASE WHEN obesity_level_code >= 5 THEN 1 ELSE 0 END) AS obesity_count,
    SUM(CASE WHEN obesity_level_code IN (3, 4) THEN 1 ELSE 0 END) AS overweight_count,
    SUM(CASE WHEN obesity_level_code = 2 THEN 1 ELSE 0 END) AS normal_count,
    SUM(CASE WHEN obesity_level_code = 1 THEN 1 ELSE 0 END) AS underweight_count,
    ROUND(SUM(CASE WHEN obesity_level_code >= 5 THEN 1 ELSE 0 END) * 100.0 / COUNT(*), 2) AS obesity_rate,
    ROUND(SUM(CASE WHEN obesity_level_code IN (3, 4) THEN 1 ELSE 0 END) * 100.0 / COUNT(*), 2) AS overweight_rate,
    ROUND(AVG(bmi), 2) AS avg_bmi,
    ROUND(AVG(risk_score), 2) AS avg_risk_score
FROM dwd_obesity_risk_fact
WHERE dt='2026-02-02'
GROUP BY gender;

3.2.4 ADS层应用数据

肥胖趋势分析：

INSERT OVERWRITE TABLE ads_obesity_trend_analysis PARTITION (dt='2026-02-02')
SELECT 
    age_group,
    gender,
    total_count,
    obesity_count,
    obesity_rate,
    avg_bmi,
    avg_risk_score
FROM dws_gender_obesity
WHERE dt='2026-02-02';

实现说明：

1. 使用单独的INSERT语句替代UNION ALL
2. 支持按日期分区，便于数据管理和查询
3. 启用Hive向量化执行，提升查询性能
4. 启用并行执行，充分利用集群资源

3.3 后端API实现

3.3.1 Controller层

ObesityAnalysisController.java 提供RESTful API接口：

@RestController
@RequestMapping("/api/obesity")
public class ObesityAnalysisController {

    @Resource
    private ObesityAnalysisService obesityAnalysisService;

    @GetMapping("/trend")
    public List<AdsObesityTrendAnalysis> getObesityTrend() {
        return obesityAnalysisService.getObesityTrendAnalysis();
    }

    @GetMapping("/trend/page")
    public Object getObesityTrendPage(@RequestParam(defaultValue = "1") int current, 
                                      @RequestParam(defaultValue = "10") int size) {
        Page<AdsObesityTrendAnalysis> page = new Page<>(current, size);
        return obesityAnalysisService.getObesityTrendAnalysisPage(page);
    }

    @GetMapping("/risk-factors")
    public List<AdsObesityRiskFactors> getRiskFactors() {
        return obesityAnalysisService.getObesityRiskFactors();
    }

    @GetMapping("/demographic")
    public List<AdsObesityDemographicAnalysis> getDemographicAnalysis() {
        return obesityAnalysisService.getDemographicAnalysis();
    }

    @GetMapping("/lifestyle")
    public List<AdsObesityLifestyleAnalysis> getLifestyleAnalysis() {
        return obesityAnalysisService.getLifestyleAnalysis();
    }

    @GetMapping("/comprehensive")
    public List<AdsObesityComprehensiveAnalysis> getComprehensiveAnalysis() {
        return obesityAnalysisService.getComprehensiveAnalysis();
    }

    @GetMapping("/raw")
    public List<OdsObesityRaw> getRawObesityData() {
        return obesityAnalysisService.getRawObesityData();
    }

    @GetMapping("/health")
    public String health() {
        return "Obesity Analysis API is healthy!";
    }
}

3.3.2 Service层

ObesityAnalysisServiceImpl.java 实现业务逻辑：

@Service
public class ObesityAnalysisServiceImpl implements ObesityAnalysisService {
    
    @Resource
    private AdsObesityTrendAnalysisMapper adsObesityTrendAnalysisMapper;
    
    @Resource
    private AdsObesityRiskFactorsMapper adsObesityRiskFactorsMapper;
    
    @Resource
    private AdsObesityDemographicAnalysisMapper adsObesityDemographicAnalysisMapper;
    
    @Resource
    private AdsObesityLifestyleAnalysisMapper adsObesityLifestyleAnalysisMapper;
    
    @Resource
    private AdsObesityComprehensiveAnalysisMapper adsObesityComprehensiveAnalysisMapper;
    
    @Resource
    private OdsObesityRawMapper odsObesityRawMapper;
    
    @Override
    public List<AdsObesityTrendAnalysis> getObesityTrendAnalysis() {
        return adsObesityTrendAnalysisMapper.selectList(null);
    }
    
    @Override
    public Page<AdsObesityTrendAnalysis> getObesityTrendAnalysisPage(Page<AdsObesityTrendAnalysis> page) {
        return adsObesityTrendAnalysisMapper.selectPage(page, null);
    }
    
    @Override
    public List<AdsObesityRiskFactors> getObesityRiskFactors() {
        return adsObesityRiskFactorsMapper.selectList(null);
    }
    
    @Override
    public List<AdsObesityDemographicAnalysis> getDemographicAnalysis() {
        return adsObesityDemographicAnalysisMapper.selectList(null);
    }
    
    @Override
    public List<AdsObesityLifestyleAnalysis> getLifestyleAnalysis() {
        return adsObesityLifestyleAnalysisMapper.selectList(null);
    }
    
    @Override
    public List<AdsObesityComprehensiveAnalysis> getComprehensiveAnalysis() {
        return adsObesityComprehensiveAnalysisMapper.selectList(null);
    }
    
    @Override
    public List<OdsObesityRaw> getRawObesityData() {
        return odsObesityRawMapper.selectList(null);
    }
    
    @Override
    public Page<OdsObesityRaw> getRawObesityDataPage(Page<OdsObesityRaw> page) {
        return odsObesityRawMapper.selectPage(page, null);
    }
}

3.3.3 Entity层

AdsObesityTrendAnalysis.java 实体类：

@Data
@TableName("ads_obesity_trend_analysis")
public class AdsObesityTrendAnalysis {
    @TableId(type = IdType.AUTO)
    private Long id;
    private String ageGroup;
    private String gender;
    private Long totalCount;
    private Long obesityCount;
    private BigDecimal obesityRate;
    private BigDecimal avgBmi;
    private BigDecimal avgRiskScore;
    private String dt;
    private LocalDateTime createTime;
    private LocalDateTime updateTime;
}

实现说明：

1. RESTful API设计：遵循REST规范
2. 分页查询：使用MyBatis-Plus的Page对象实现分页
3. 依赖注入：使用@Resource注解进行依赖注入
4. ORM映射：使用MyBatis-Plus简化数据库操作

3.4 前端可视化实现

3.4.1 Dashboard页面

Dashboard.vue 主页面展示关键指标和图表：

<template>
  <div class="dashboard-container">
    <el-card class="welcome-card">
      <h2>肥胖风险分析系统</h2>
      <p>欢迎使用肥胖风险分析系统，这里展示了肥胖趋势、人口统计、生活方式等多维度的分析数据。</p>
    </el-card>

    <div class="stats-container">
      <el-card class="stat-card">
        <div class="stat-icon total"><i class="el-icon-user"></i></div>
        <div class="stat-info">
          <h3>总样本数</h3>
          <p class="stat-value">{{ totalSamples }}</p>
        </div>
      </el-card>
      <el-card class="stat-card">
        <div class="stat-icon obesity"><i class="el-icon-warning"></i></div>
        <div class="stat-info">
          <h3>肥胖人数</h3>
          <p class="stat-value">{{ obesityCount }}</p>
        </div>
      </el-card>
      <el-card class="stat-card">
        <div class="stat-icon rate"><i class="el-icon-data-line"></i></div>
        <div class="stat-info">
          <h3>肥胖率</h3>
          <p class="stat-value">{{ obesityRate }}%</p>
        </div>
      </el-card>
      <el-card class="stat-card">
        <div class="stat-icon bmi"><i class="el-icon-s-data"></i></div>
        <div class="stat-info">
          <h3>平均BMI</h3>
          <p class="stat-value">{{ avgBmi }}</p>
        </div>
      </el-card>
    </div>

    <div class="charts-container">
      <div class="chart-row">
        <el-card class="chart-card">
          <div slot="header" class="chart-header">
            <span>肥胖趋势分析</span>
            <el-button type="primary" size="small" @click="navigateTo('/trend-analysis')">查看详情</el-button>
          </div>
          <div ref="trendChart" class="chart"></div>
        </el-card>
        <el-card class="chart-card">
          <div slot="header" class="chart-header">
            <span>风险因素分析</span>
            <el-button type="primary" size="small" @click="navigateTo('/risk-factors-analysis')">查看详情</el-button>
          </div>
          <div ref="riskFactorsChart" class="chart"></div>
        </el-card>
      </div>
    </div>
  </div>
</template>

3.4.2 Vuex状态管理

store/index.js 管理全局状态：

export default new Vuex.Store({
  state: {
    loading: false,
    obesityTrendData: [],
    riskFactorsData: [],
    demographicData: [],
    lifestyleData: [],
    comprehensiveData: []
  },
  getters: {
    getObesityTrendData: state => state.obesityTrendData,
    getRiskFactorsData: state => state.riskFactorsData,
    getDemographicData: state => state.demographicData,
    getLifestyleData: state => state.lifestyleData,
    getComprehensiveData: state => state.comprehensiveData
  },
  mutations: {
    setLoading(state, status) {
      state.loading = status
    },
    setObesityTrendData(state, data) {
      state.obesityTrendData = data
    },
    setRiskFactorsData(state, data) {
      state.riskFactorsData = data
    },
    setDemographicData(state, data) {
      state.demographicData = data
    },
    setLifestyleData(state, data) {
      state.lifestyleData = data
    },
    setComprehensiveData(state, data) {
      state.comprehensiveData = data
    }
  },
  actions: {
    fetchObesityTrendData({ commit }) {
      commit('setLoading', true)
      return new Promise((resolve, reject) => {
        fetch('http://localhost:8080/api/api/obesity/trend')
          .then(response => response.json())
          .then(data => {
            commit('setObesityTrendData', data)
            commit('setLoading', false)
            resolve(data)
          })
          .catch(error => {
            commit('setLoading', false)
            reject(error)
          })
      })
    }
  }
})

实现说明：

1. 组件化设计：使用Vue组件化开发
2. 状态管理：使用Vuex管理全局状态
3. 数据可视化：使用ECharts实现图表展示
4. 响应式布局：适配不同屏幕尺寸

四、技术实现要点

4.1 MapReduce实现要点

1. HDFS连接检测：检测HDFS可用性，失败时使用本地文件系统
2. 数据质量监控：使用Hadoop Counter统计数据质量指标
3. 衍生字段计算：在Map阶段计算BMI、风险评分等衍生字段
4. 数据标准化：对性别、肥胖水平等字段进行标准化处理

4.2 Hive ETL实现要点

1. 移除UNION ALL：使用单独的INSERT语句
2. 并行执行：启用并行执行，充分利用集群资源
3. 向量化执行：启用Hive向量化执行，提升查询性能
4. 动态分区：支持按日期分区，便于数据管理和查询
5. 小文件合并：配置小文件合并参数，减少NameNode压力

4.3 后端API实现要点

1. RESTful设计：遵循REST规范
2. 分页查询：使用MyBatis-Plus的Page对象实现分页
3. 依赖注入：使用Spring的依赖注入
4. ORM映射：使用MyBatis-Plus简化数据库操作

4.4 前端实现要点

1. 组件化开发：使用Vue组件化开发
2. 状态管理：使用Vuex管理全局状态
3. 数据可视化：使用ECharts实现图表展示
4. 响应式布局：适配不同屏幕尺寸

五、项目部署

5.1 环境要求

• JDK 1.8+
• Hadoop 3.x
• Hive 3.x
• MySQL 5.7+
• Node.js 14+
• Maven 3.6+

5.2 部署步骤

1. 配置Hadoop环境
2. 配置Hive环境
3. 编译MapReduce程序
4. 执行MapReduce数据清洗
5. 执行Hive建表和ETL脚本
6. 配置Sqoop数据导出
7. 启动Spring Boot后端服务
8. 启动Vue前端服务

六、总结

本文介绍了基于Hadoop、Hive、SpringBoot、Vue.js等技术栈实现的肥胖风险数据分析系统。系统实现了以下功能：

1. 数据清洗：使用MapReduce对原始数据进行清洗和转换
2. 数据仓库：构建ODS/DWD/DWS/ADS四层数据仓库
3. 数据分析：实现多维度数据分析
4. 数据可视化：使用ECharts实现数据可视化展示
5. API服务：提供RESTful API接口

本文提供了主要技术实现的代码示例，供读者参考。