Python数据分析实战：学习行为特征挖掘实验

import pandas as pd
import numpy as np

读取数据
df = pd.read_csv('learning_data_6months.csv')

数据清洗
1. 处理缺失值
 df = df.dropna(subset=['user_id', 'start_time', 'duration'])

2. 转换时间格式
 df['start_time'] = pd.to_datetime(df['start_time'])
 df['hour'] = df['start_time'].dt.hour
 df['weekday'] = df['start_time'].dt.weekday

3. 过滤异常值（学习时长<10秒或>3600秒）
 df = df[(df['duration'] >= 10) & (df['duration'] <= 3600)]

import matplotlib.pyplot as plt

设置中文字体
plt.rcParams['font.sans-serif'] = ['WenQuanYi Zen Hei']

按小时统计学习次数
hour_dist = df.groupby('hour').size()

绘制学习时段分布图
plt.figure(figsize=(12, 6))
hour_dist.plot(kind='bar', color='#165DFF')
plt.title('用户学习时段分布', fontsize=14)
plt.xlabel('小时', fontsize=12)
plt.ylabel('学习次数', fontsize=12)
plt.xticks(rotation=0)
plt.grid(axis='y', alpha=0.3)
plt.savefig('learning_hour_dist.png', dpi=300, bbox_inches='tight')
plt.close()

from sklearn.cluster import KMeans
from sklearn.preprocessing import StandardScaler

提取特征变量
features = df[['duration', 'completion_rate', 'hour', 'weekday']]

数据标准化
scaler = StandardScaler()
features_scaled = scaler.fit_transform(features)

K-Means聚类（分为3类：高效/中等/低效）
kmeans = KMeans(n_clusters=3, random_state=42)
df['efficiency_type'] = kmeans.fit_predict(features_scaled)

分析各类别特征
cluster_analysis = df.groupby('efficiency_type').agg({
    'duration': 'mean',
    'completion_rate': 'mean',
    'hour': 'mean',
    'user_id': 'nunique'
}).round(2)

print(cluster_analysis)