데이터분석가 취준생을 위한 필수 데이터분석 기법 5가지와 실전 예제

import pandas as pd
import numpy as np

# 예시: 이커머스 주문 데이터
data = pd.DataFrame({
    '주문금액': [15000, 23000, 45000, 12000, 67000, 34000, 18000],
    '배송시간': [2, 3, 1, 4, 2, 3, 2]
})

# 기술통계 요약
print(data.describe())

# 출력:
#        주문금액      배송시간
# count    7.0       7.0
# mean   30571.4     2.43
# std    19234.5     0.98
# min    12000.0     1.0
# 25%    16500.0     2.0
# 50%    23000.0     2.0
# 75%    39500.0     3.0
# max    67000.0     4.0

import seaborn as sns
import matplotlib.pyplot as plt

# 예시: 마케팅 캠페인 데이터
marketing_data = pd.DataFrame({
    '광고비': [100, 150, 200, 250, 300],
    '매출': [500, 650, 800, 950, 1100],
    '방문자수': [1000, 1300, 1600, 1900, 2200]
})

# 상관계수 행렬
corr_matrix = marketing_data.corr()
print(corr_matrix)

# 히트맵 시각화
sns.heatmap(corr_matrix, annot=True, cmap='coolwarm')
plt.show()

from sklearn.linear_model import LinearRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error, r2_score

# 예시: 부동산 가격 예측
house_data = pd.DataFrame({
    '평수': [20, 25, 30, 35, 40, 45, 50],
    '역세권': [1, 1, 0, 1, 0, 1, 1],  # 1=역세권, 0=비역세권
    '가격': [3.0, 3.5, 3.2, 4.0, 3.8, 4.5, 5.0]  # 억 원
})

X = house_data[['평수', '역세권']]
y = house_data['가격']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)

# 모델 학습
model = LinearRegression()
model.fit(X_train, y_train)

# 예측 및 평가
y_pred = model.predict(X_test)
print(f"R² Score: {r2_score(y_test, y_pred):.3f}")
print(f"평수 계수: {model.coef_[0]:.3f}억/평")
print(f"역세권 계수: {model.coef_[1]:.3f}억")

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

# 예시: 고객 세분화
customer_data = pd.DataFrame({
    '월평균구매액': [10, 50, 100, 15, 80, 120, 20, 90],
    '구매빈도': [2, 8, 15, 3, 12, 18, 4, 14]
})

# 정규화
scaler = StandardScaler()
scaled_data = scaler.fit_transform(customer_data)

# K-Means 군집화 (3개 그룹)
kmeans = KMeans(n_clusters=3, random_state=42)
customer_data['군집'] = kmeans.fit_predict(scaled_data)

print(customer_data.groupby('군집').mean())

from statsmodels.tsa.seasonal import seasonal_decompose
import matplotlib.pyplot as plt

# 예시: 월별 웹사이트 방문자 수
dates = pd.date_range('2024-01', periods=12, freq='M')
visitors = [1000, 1100, 1300, 1250, 1400, 1600, 1800, 1750, 1650, 1550, 1700, 2000]

ts_data = pd.Series(visitors, index=dates)

# 시계열 분해
decomposition = seasonal_decompose(ts_data, model='additive', period=4)
decomposition.plot()
plt.show()