✨作者主页:IT研究室✨
个人简介:曾从事计算机专业培训教学,擅长Java、Python、微信小程序、Golang、安卓Android等项目实战。接项目定制开发、代码讲解、答辩教学、文档编写、降重等。
☑文末获取源码☑
精彩专栏推荐⬇⬇⬇
Java项目
Python项目
安卓项目
微信小程序项目
一、前言
系统介绍
本系统名为《基于大数据的睡前手机使用行为与睡眠关联可视化分析系统》,面向计算机专业本科毕业设计场景,围绕睡前手机使用行为与睡眠质量之间的关联关系展开数据化分析。系统以Hadoop HDFS作为底层数据存储,借助Spark与Spark SQL完成睡前手机行为数据、睡眠数据的清洗、聚合与关联计算,并结合Pandas、NumPy做指标加工,后端采用Django或Spring Boot提供数据接口,前端基于Vue、ElementUI与Echarts构建可视化大屏。功能上包含手机行为分析、睡眠质量分析、关联洞察分析三大分析模块及对应维度分析,配套睡前手机睡眠数据管理、数据大屏展示、个人信息与账户卷宗、修改密码等基础管理能力,能够把零散的睡前用机记录与睡眠记录整理成可读的维度指标和关联结论,帮助使用者直观看到睡前手机使用与睡眠之间的对应关系。
选题背景
睡前刷手机已经成了很多人的日常习惯,手机使用时长、使用时段与入睡时间、睡眠时长之间往往存在某种对应关系,但这种关系大多停留在个人感受层面,缺少系统化的数据整理与分析手段。计算机专业毕业设计阶段,不少同学希望做一个既能体现大数据技术、又贴近真实生活场景的题目,睡前手机使用与睡眠数据恰好具备记录量大、维度多、适合做关联分析的特点。传统做法多是用表格手工统计,数据量稍大就难以支撑多维度的对比分析,也难以把行为数据与睡眠数据放在同一套指标下观察。基于Hadoop与Spark的大数据技术栈,可以对睡前手机行为记录和睡眠记录进行分布式存储与并行计算,再通过可视化方式把分析结果呈现出来,这为本课题的开展提供了技术上的可行性,也让选题具备了真实的数据分析对象。
选题意义
本课题的意义更多体现在实际动手层面。对即将毕业的计算机专业学生来说,它提供了一个把Hadoop、Spark、Spark SQL与可视化技术串起来练手的完整场景,从数据存储、清洗聚合到指标计算、图表呈现,能较完整地走一遍大数据分析的流程,对理解大数据技术栈的实际用法有帮助。对使用者而言,系统把睡前手机使用行为和睡眠数据放在一起做维度分析和关联洞察,能把原本分散的记录整理成看得懂的指标和大屏图表,方便观察睡前用机与睡眠之间的对应情况,具备一定的参考价值。从毕业设计本身看,课题贴近生活、数据对象清晰、技术路线明确,规模适中,适合作为本科阶段的综合实践题目,也能为后续类似的数据分析类课题提供一个可复用的实现思路。整体来说,它是一个偏实践、偏应用的小型系统,意义在于把课堂所学落到一个具体可做的题目上。
二、开发环境
- 大数据框架:Hadoop+Spark(本次没用Hive,支持定制)
- 开发语言:Python+Java(两个版本都支持)
- 后端框架:Django+Spring Boot(Spring+SpringMVC+Mybatis)(两个版本都支持)
- 前端:Vue+ElementUI+Echarts+HTML+CSS+JavaScript+jQuery
- 详细技术点:Hadoop、HDFS、Spark、Spark SQL、Pandas、NumPy
- 数据库:MySQL
三、系统界面展示
- 基于大数据的睡前手机使用行为与睡眠关联性可视化分析系统界面展示:








四、代码参考
- 项目实战代码参考:
from pyspark.sql import SparkSession
from pyspark.sql import functions as F
from django.http import JsonResponse
from django.views.decorators.csrf import csrf_exempt
import json
spark = SparkSession.builder.appName("SleepPhoneAnalysis").master("local[*]").config("spark.sql.shuffle.partitions", "4").enableHiveSupport().getOrCreate()
def phone_behavior_analysis(request):
params = json.loads(request.body)
user_id = params.get("userId")
start_date = params.get("startDate")
end_date = params.get("endDate")
behavior_df = spark.read.parquet("hdfs://localhost:9000/sleep_phone/phone_behavior")
filtered_df = behavior_df.filter((F.col("user_id") == user_id) & (F.col("record_date") >= start_date) & (F.col("record_date") <= end_date))
filtered_df.createOrReplaceTempView("phone_behavior_view")
duration_df = spark.sql("SELECT record_date, SUM(use_duration) AS total_duration, COUNT(1) AS use_times, AVG(use_duration) AS avg_duration FROM phone_behavior_view GROUP BY record_date ORDER BY record_date")
app_type_df = spark.sql("SELECT app_type, SUM(use_duration) AS type_duration, COUNT(1) AS type_times FROM phone_behavior_view GROUP BY app_type ORDER BY type_duration DESC")
hour_df = spark.sql("SELECT use_hour, SUM(use_duration) AS hour_duration FROM phone_behavior_view GROUP BY use_hour ORDER BY use_hour")
total_duration = duration_df.agg(F.sum("total_duration")).collect()[0][0] or 0
avg_daily_duration = duration_df.agg(F.avg("total_duration")).collect()[0][0] or 0
late_night_df = filtered_df.filter(F.col("use_hour") >= 23)
late_night_duration = late_night_df.agg(F.sum("use_duration")).collect()[0][0] or 0
late_night_ratio = round(late_night_duration / total_duration, 4) if total_duration > 0 else 0
duration_list = [{"date": row["record_date"], "totalDuration": row["total_duration"], "useTimes": row["use_times"], "avgDuration": round(row["avg_duration"], 2)} for row in duration_df.collect()]
app_type_list = [{"appType": row["app_type"], "typeDuration": row["type_duration"], "typeTimes": row["type_times"]} for row in app_type_df.collect()]
hour_list = [{"useHour": row["use_hour"], "hourDuration": row["hour_duration"]} for row in hour_df.collect()]
result = {"userId": user_id, "totalDuration": total_duration, "avgDailyDuration": round(avg_daily_duration, 2), "lateNightRatio": late_night_ratio, "durationTrend": duration_list, "appTypeDistribution": app_type_list, "hourDistribution": hour_list}
return JsonResponse({"code": 200, "msg": "手机行为分析完成", "data": result})
def sleep_quality_analysis(request):
params = json.loads(request.body)
user_id = params.get("userId")
start_date = params.get("startDate")
end_date = params.get("endDate")
sleep_df = spark.read.parquet("hdfs://localhost:9000/sleep_phone/sleep_record")
filtered_df = sleep_df.filter((F.col("user_id") == user_id) & (F.col("record_date") >= start_date) & (F.col("record_date") <= end_date))
filtered_df.createOrReplaceTempView("sleep_record_view")
daily_df = spark.sql("SELECT record_date, sleep_duration, sleep_score, fall_asleep_time, wake_up_time FROM sleep_record_view ORDER BY record_date")
score_df = spark.sql("SELECT AVG(sleep_score) AS avg_score, MAX(sleep_score) AS max_score, MIN(sleep_score) AS min_score, AVG(sleep_duration) AS avg_sleep_duration FROM sleep_record_view")
score_row = score_df.collect()[0]
avg_score = round(score_row["avg_score"], 2) if score_row["avg_score"] else 0
max_score = score_row["max_score"] or 0
min_score = score_row["min_score"] or 0
avg_sleep_duration = round(score_row["avg_sleep_duration"], 2) if score_row["avg_sleep_duration"] else 0
quality_level_df = filtered_df.withColumn("quality_level", F.when(F.col("sleep_score") >= 85, "优秀").when(F.col("sleep_score") >= 70, "良好").when(F.col("sleep_score") >= 60, "一般").otherwise("较差"))
quality_count_df = quality_level_df.groupBy("quality_level").agg(F.count(1).alias("level_count"))
quality_list = [{"qualityLevel": row["quality_level"], "levelCount": row["level_count"]} for row in quality_count_df.collect()]
duration_list = [{"date": row["record_date"], "sleepDuration": row["sleep_duration"], "sleepScore": row["sleep_score"], "fallAsleepTime": row["fall_asleep_time"], "wakeUpTime": row["wake_up_time"]} for row in daily_df.collect()]
good_days = filtered_df.filter(F.col("sleep_score") >= 70).count()
total_days = filtered_df.count()
good_ratio = round(good_days / total_days, 4) if total_days > 0 else 0
result = {"userId": user_id, "avgScore": avg_score, "maxScore": max_score, "minScore": min_score, "avgSleepDuration": avg_sleep_duration, "goodRatio": good_ratio, "qualityDistribution": quality_list, "sleepTrend": duration_list}
return JsonResponse({"code": 200, "msg": "睡眠质量分析完成", "data": result})
def correlation_insight_analysis(request):
params = json.loads(request.body)
user_id = params.get("userId")
start_date = params.get("startDate")
end_date = params.get("endDate")
behavior_df = spark.read.parquet("hdfs://localhost:9000/sleep_phone/phone_behavior")
sleep_df = spark.read.parquet("hdfs://localhost:9000/sleep_phone/sleep_record")
behavior_filtered = behavior_df.filter((F.col("user_id") == user_id) & (F.col("record_date") >= start_date) & (F.col("record_date") <= end_date))
sleep_filtered = sleep_df.filter((F.col("user_id") == user_id) & (F.col("record_date") >= start_date) & (F.col("record_date") <= end_date))
behavior_daily = behavior_filtered.groupBy("record_date").agg(F.sum("use_duration").alias("total_duration"), F.sum(F.when(F.col("use_hour") >= 23, F.col("use_duration")).otherwise(0)).alias("late_night_duration"))
sleep_daily = sleep_filtered.groupBy("record_date").agg(F.avg("sleep_score").alias("avg_sleep_score"), F.avg("sleep_duration").alias("avg_sleep_duration"), F.avg("fall_asleep_time").alias("avg_fall_asleep_time"))
joined_df = behavior_daily.join(sleep_daily, on="record_date", how="inner")
joined_df.createOrReplaceTempView("joined_view")
corr_df = spark.sql("SELECT corr(total_duration, avg_sleep_score) AS duration_score_corr, corr(late_night_duration, avg_sleep_score) AS late_score_corr, corr(total_duration, avg_fall_asleep_time) AS duration_fall_corr FROM joined_view")
corr_row = corr_df.collect()[0]
duration_score_corr = round(corr_row["duration_score_corr"], 4) if corr_row["duration_score_corr"] else 0
late_score_corr = round(corr_row["late_score_corr"], 4) if corr_row["late_score_corr"] else 0
duration_fall_corr = round(corr_row["duration_fall_corr"], 4) if corr_row["duration_fall_corr"] else 0
group_df = joined_df.withColumn("duration_level", F.when(F.col("total_duration") >= 240, "高时长").when(F.col("total_duration") >= 120, "中时长").otherwise("低时长"))
group_stat_df = group_df.groupBy("duration_level").agg(F.avg("avg_sleep_score").alias("group_sleep_score"), F.avg("avg_fall_asleep_time").alias("group_fall_asleep_time"), F.count(1).alias("group_days"))
group_list = [{"durationLevel": row["duration_level"], "groupSleepScore": round(row["group_sleep_score"], 2), "groupFallAsleepTime": round(row["group_fall_asleep_time"], 2), "groupDays": row["group_days"]} for row in group_stat_df.collect()]
scatter_list = [{"date": row["record_date"], "totalDuration": row["total_duration"], "lateNightDuration": row["late_night_duration"], "avgSleepScore": round(row["avg_sleep_score"], 2), "avgSleepDuration": round(row["avg_sleep_duration"], 2)} for row in joined_df.collect()]
if late_score_corr < -0.3:
insight_text = "睡前深夜用机时长与睡眠得分呈较明显负相关,深夜用机越久睡眠得分偏低"
elif late_score_corr < 0:
insight_text = "睡前深夜用机时长与睡眠得分呈轻微负相关,存在一定影响"
else:
insight_text = "当前数据下睡前用机与睡眠得分未呈现明显负相关"
result = {"userId": user_id, "durationScoreCorr": duration_score_corr, "lateScoreCorr": late_score_corr, "durationFallCorr": duration_fall_corr, "groupStat": group_list, "scatterData": scatter_list, "insightText": insight_text}
return JsonResponse({"code": 200, "msg": "关联洞察分析完成", "data": result})
五、系统视频
基于大数据的睡前手机使用行为与睡眠关联性可视化分析系统项目视频:
演示视频
结语
最新大数据毕业设计选题推荐-基于大数据的睡前手机使用行为与睡眠关联性可视化分析系统-大数据-Spark-Hadoop-Bigdata
想看其他类型的计算机毕业设计作品也可以和我说都有 谢谢大家!
有技术这一块问题大家可以评论区交流或者私我~
大家可以帮忙点赞、收藏、关注、评论啦~
源码获取:⬇⬇⬇
391



被折叠的 条评论
为什么被折叠?



