题解 | #某店铺用户消费特征评分#

import pandas as pd

df = pd.read_csv("sales.csv")

# 设定规则
q1_recency = df["recency"].quantile(0.25)
median_recency = df["recency"].quantile(0.5)
q3_recency = df["recency"].quantile(0.75)

q1_frequency = df["frequency"].quantile(0.25)
median_frequency = df["frequency"].quantile(0.5)
q3_frequency = df["frequency"].quantile(0.75)

q1_monetary = df["monetary"].quantile(0.25)
median_monetary = df["monetary"].quantile(0.5)
q3_monetary = df["monetary"].quantile(0.75)

# 为 recency 特征评分
def score_recency(value):
    if value <= q1_recency:
        return 4
    elif value <= median_recency:
        return 3
    elif value <= q3_recency:
        return 2
    else:
        1


# 为 frequency 特征评分
def score_frequency(value):
    if value <= q1_frequency:
        return 1
    elif value <= median_frequency:
        return 2
    elif value <= q3_frequency:
        return 3
    else:
        4


# 为 monetary 特征评分
def score_monetary(value):
    if value <= q1_monetary:
        return 1
    elif value <= median_monetary:
        return 2
    elif value <= q3_monetary:
        return 3
    else:
        4

df['R_Quartile'] = df['recency'].apply(score_recency)
df['F_Quartile'] = df['frequency'].apply(score_frequency)
df['M_Quartile'] = df['monetary'].apply(score_monetary)

print(df.head(5))