人工智能|机器学习——基于机器学习的信用卡办卡意愿模型预测项目-CFANZ编程社区

一、背景介绍

二、数据准备

# 数据库连接和数据获取
import pandas as pd
import pymysql
from data.mapper import host, user, password, database

# 连接MySQL数据库
conn = pymysql.connect(
    host=host,
    user=user,
    password=password,
    database=database
)

# 从MySQL数据库中读取处理后的数据
query = "SELECT * FROM processed_customer_data"
df = pd.read_sql(query, conn)

# 关闭数据库连接
conn.close()

三、模型训练与评估

3.1 随机森林模型

# 随机森林模型训练与评估
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, classification_report, confusion_matrix

# 特征与标签分割
X = df.drop(columns=['Attrition_Flag'])
y = df['Attrition_Flag']

# 数据集划分
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=2)

# 随机森林模型训练
rf_model = RandomForestClassifier()
rf_model.fit(X_train, y_train)

# 模型预测
y_pred = rf_model.predict(X_test)

# 模型评估
accuracy = accuracy_score(y_test, y_pred)
classification_rep = classification_report(y_test, y_pred)
conf_matrix = confusion_matrix(y_test, y_pred)

3.2 逻辑回归模型

# 逻辑回归模型训练与评估
from sklearn.linear_model import LogisticRegression

# 数据集划分
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 逻辑回归模型训练
logreg_model = LogisticRegression()
logreg_model.fit(X_train, y_train)

# 模型预测
y_pred = logreg_model.predict(X_test)

# 模型评估
accuracy = accuracy_score(y_test, y_pred)
classification_rep = classification_report(y_test, y_pred)
conf_matrix = confusion_matrix(y_test, y_pred)

四、数据可视化

# Django视图函数中的数据可视化
from django.shortcuts import render
from pyecharts.charts import Bar, Pie, Line
from pyecharts import options as opts
from pyecharts.globals import CurrentConfig, ThemeType

from web.service.task_service import get_custormer_age, get_income_category, get_education_level, get_credit_limit, \
    get_months_inactive_12_mon

def bar_chart(request):
    # 获取客户年龄分布数据
    x, y = get_custormer_age()
    line = (
        Line()
        .add_xaxis([str(age) for age in x])
        .add_yaxis("Count", y)
        .set_global_opts(
            title_opts=opts.TitleOpts(title="客户年龄分布图"),
            xaxis_opts=opts.AxisOpts(name="Age"),
            yaxis_opts=opts.AxisOpts(name="Count"),
        )
    )

    # 获取客户信用卡额度分布数据
    x1, y1 = get_credit_limit()
    line1 = (
        Line()
        .add_xaxis([str(age) for age in x1])
        .add_yaxis("Count", y1)
        .set_global_opts(
            title_opts=opts.TitleOpts(title="客户信用卡额度top10分布图"),
            xaxis_opts=opts.AxisOpts(name="Age"),
            yaxis_opts=opts.AxisOpts(name="Count"),
        )
    )

    # 获取客户非活跃月数分布数据
    bar1 = Bar()
    x1, y1 = get_months_inactive_12_mon()
    bar1.add_xaxis(x1)
    bar1.add_yaxis("客户去年非活跃月数分布", y1)

    # 获取客户收入范围趋势数据
    bar = Bar()
    x, y = get_income_category()
    bar.add_xaxis(x)
    bar.add_yaxis("收入范围趋势图", y)

    # 获取客户教育水平分布数据
    pie = Pie()
    tuple = get_education_level()
    pie.add("教育水平分布图", tuple)

    # 获取图表的JavaScript代码
    line_js = line.render_embed()
    bar_js = bar.render_embed()
    pie_js = pie.render_embed()
    bar1_js = bar1.render_embed()
    line1_js = line1.render_embed()

    return render(request, 'charts/bar_chart.html', {'line': line_js, 'bar': bar_js, 'pie': pie_js, 'line1': line1_js, 'bar1': bar1_js})