Ecommerce Machine Learning - 광고 반응률 예측(로지스틱 회귀)

데이터 확인

import pandas as pd
import numpy as np

import matplotlib.pyplot as plt
import seaborn as sns

data = pd.read_csv('advertising.csv')
data
data.info()

sns.distplot(data['Area Income'])

sns.distplot(data['Age'])

# 텍스트로 된 컬럼 확인
data['Country'].nunique()        #   237
data['City'].nunique()           #   969
data['Ad Topic Line'].nunique()  #   1000

 

결측값 확인 및 처리

# 결측값 확인 및 처리
data.isna().sum() / len(data)

data.dropna()
data.drop('Age', axis=1)
round(data['Age'].mean())   # 36
data['Age'].median()        # 35.0

# 결측값 평균으로 대체
data = data.fillna(round(data['Age'].mean()))

data.isna().sum()

모델링

# train / test 나누기
from sklearn.model_selection import train_test_split
X = data[['Daily Time Spent on Site','Age','Area Income', 'Daily Internet Usage','Male']]
y = data['Clicked on Ad']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2, random_state = 100)

# 로지스틱 리그레션 모델 만들기
from sklearn.linear_model import LogisticRegression
model = LogisticRegression()
model.fit(X_train, y_train)

# Coefficient 확인
model.coef_

# 예측 및 평가
pred = model.predict(X_test)
y_test

from sklearn.metrics import accuracy_score, confusion_matrix
accuracy_score(y_test, pred)            # 0.9
confusion_matrix(y_test, pred)          # 아래 사진 참조