利用python分析电商数据

私域电商私域电商 2022-11-16 15:55 6248人围观

操纵python分析电商数据

import pandas as pd
data = pd.read_csv('/Users/jenifferli_0605hotmail.com/Downloads/data.csv',encoding='ISO-8859-1',nrows=541910)
data.head()

看下数据集

data.shape

检察数据集

data.info()

检察缺失值以及缺失值占比

data.isnull().sum()

missing_percentage = data.isnull().sum() / data.shape[0] * 100
missing_percentage

删除缺失值

data1 = data.dropna(how='any')
data1.info()

日期范例转换

data1['InvoiceDate'] = pd.to_datetime(data1['InvoiceDate']) #12/1/2010 8:26转换为2010-12-01
data1['InvoiceDate'] =data1['InvoiceDate'].APPly(lambda x: x.strftime('%Y-%m-%d')) #将datetime64[ns] 转为 object
data1.info()

新建一个销售金额的字段，即产物数目乘以单价。

data1['Price'] =data1['Quantity']*data1['UnitPrice']
data1.head()

【第一个题目：采办商品数前十的国家是？】

data1[data1['Quantity']>0].groupby('Country')['Quantity'].sum().sort_values(ascending=False).head(10)

可视化

plt.subplot(1,1,1)
x=np.array(["UK","NLD","EIRE","GER ","France","AUS","SE","Swi","Spain ","Jpn "])
y=np.array([4269472,2009370,140525,119263,111472,84209,36083,30083,27951,26016])
plt.bar(x,y,width = 0.5,align = "center",label = "采办量",color='green')
plt.title("采办商品数前十的国家",loc = "center")
for a,b in zip (x,y):
plt.text(a,b,b,ha='center',va= 'bottom',fontsize = 12)
plt.xlabel('国家')
plt.ylabel('采办量')
plt.legend()#显现图例
plt.show()

【第二个题目：买卖额前十的国家是？】

price_first_10 = data1[data1['Quantity'] > 0].groupby('Country').sum()['Price'].sort_values(ascending=False).head(10)
price_first_10.head(10)

可视化

plt.figure(figsize=(9,5), dpi=70)
plt.style.use('Solarize_Light2')
plt.bar(price_first_10.index.tolist(), price_first_10.values.tolist(), color='green', label='Quantity', alpha=0.7)
plt.title('Country Sales Top 10', pad=10)
plt.xlabel('Country')
plt.legend(loc='best')
plt.ylabel('price')
plt.xticks(rotation=45)

【第三个题目：前十国家月营业额趋向？】

data1['month'] = pd.to_datetime(data1['InvoiceDate']).dt.month
data1['year'] = pd.to_datetime(data1['InvoiceDate']).dt.year
year_11 = data1[(data1['Quantity'] > 0) & (data1['year'] == 2011)]
pivot_year_11 = year_11.pivot_table(values='Price', index='Country', columns='month', aggfunc='sum', \
margins=True).sort_values('All',ascending=False).head(11)
country_price = pivot_year_11.values.tolist()
month = pivot_year_11.columns.tolist()[:-1]
labels = pivot_year_11.index.tolist()
pivot_year_11

可视化

plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
plt.figure(figsize=(12, 9))
for data, l in zip(country_price, labels):
plt.plot(month, data[:12], 'o-', label=l)
plt.legend(loc='best')
plt.xticks([1,2,3,4,5,6,7,8,9,10,11,12])
# plt.yticks(ticks=list(np.arange(0, 12.1, 0.5)))
plt.title('营业额前十国家月支出趋向', pad=10)
plt.xlabel('月份',fontdict={'color':'black', 'fontsize':13})
plt.ylabel('营业额',fontdict={'color':'black', 'fontsize':13})

plt.show()

【第四个题目：那些月份销量较佳？】

month_first_10 = data1[data1['Quantity'] > 0].groupby('month').sum()['Quantity'].sort_values(ascending=False).head(10)
month_first_10

可视化

【第五个题目：客单价是几多？】

sumPrice=data1[data1['Quantity']>0]['Price'].sum()
countID=data1[data1['Quantity']>0]['InvoiceNo'].count()
avgPrice = sumPrice/countID
print(avgPrice)