고객 장바구니 분석 level_2

In [ ]:

In [ ]:

아이 용품과 가장 연관이 높은 물건은 무엇일까?¶

In [41]:

# https://www.kaggle.com/c/instacart-market-basket-analysis/data

In [2]:

from glob import glob
import pandas as pd

In [3]:

glob("*")

Out[3]:

['aisles.csv',
 'departments.csv',
 'orders.csv',
 'order_products__prior.csv',
 'order_products__train.csv',
 'products.csv',
 'sample_submission.csv',
 '고객 장바구니 분석 level_1.ipynb',
 '고객 장바구니 분석 level_2.ipynb']

In [4]:

pd.read_csv('products.csv').head()

Out[4]:

	product_id	product_name	aisle_id	department_id
0	1	Chocolate Sandwich Cookies	61	19
1	2	All-Seasons Salt	104	13
2	3	Robust Golden Unsweetened Oolong Tea	94	7
3	4	Smart Ones Classic Favorites Mini Rigatoni Wit...	38	1
4	5	Green Chile Anytime Sauce	5	13

In [5]:

pd.read_csv('order_products__train.csv').head()

Out[5]:

	order_id	product_id	add_to_cart_order	reordered
0	1	49302	1	1
1	1	11109	2	1
2	1	10246	3	0
3	1	49683	4	0
4	1	43633	5	1

In [6]:

pd.read_csv('order_products__prior.csv').head()

Out[6]:

	order_id	product_id	add_to_cart_order	reordered
0	2	33120	1	1
1	2	28985	2	1
2	2	9327	3	0
3	2	45918	4	1
4	2	30035	5	0

In [7]:

pd.read_csv('orders.csv').head()

Out[7]:

	order_id	user_id	eval_set	order_number	order_dow	order_hour_of_day	days_since_prior_order
0	2539329	1	prior	1	2	8	NaN
1	2398795	1	prior	2	3	7	15.0
2	473747	1	prior	3	3	12	21.0
3	2254736	1	prior	4	4	7	29.0
4	431534	1	prior	5	4	15	28.0

In [80]:

pd.read_csv('departments.csv').head(100)

Out[80]:

	department_id	department
0	1	frozen
1	2	other
2	3	bakery
3	4	produce
4	5	alcohol
5	6	international
6	7	beverages
7	8	pets
8	9	dry goods pasta
9	10	bulk
10	11	personal care
11	12	meat seafood
12	13	pantry
13	14	breakfast
14	15	canned goods
15	16	dairy eggs
16	17	household
17	18	babies
18	19	snacks
19	20	deli
20	21	missing

In [9]:

pd.read_csv('aisles.csv').head()

Out[9]:

	aisle_id	aisle
0	1	prepared soups salads
1	2	specialty cheeses
2	3	energy granola bars
3	4	instant foods
4	5	marinades meat preparation

In [89]:

aisle = pd.read_csv('aisles.csv')
pdt = pd.read_csv('products.csv')[['product_id', 'aisle_id']]
pd_ai = pd.merge(aisle, pdt)
pd_ai.head()

Out[89]:

	aisle_id	aisle	product_id
0	1	prepared soups salads	209
1	1	prepared soups salads	554
2	1	prepared soups salads	886
3	1	prepared soups salads	1600
4	1	prepared soups salads	2539

In [90]:

d_1 = pd.read_csv('order_products__train.csv')[['order_id', 'product_id', 'reordered']]
d_2 = pd.read_csv('order_products__prior.csv')[['order_id', 'product_id', 'reordered']]
order_product = pd.concat([d_1, d_2])
order_product.head()

Out[90]:

	order_id	product_id	reordered
0	1	49302	1
1	1	11109	1
2	1	10246	0
3	1	49683	0
4	1	43633	1

In [91]:

combined = pd.merge(order_product, pd_ai)[['order_id', 'aisle', 'reordered']]
combined.head()

Out[91]:

	order_id	aisle	reordered
0	1	yogurt	1
1	816049	yogurt	1
2	1242203	yogurt	1
3	1383349	yogurt	1
4	1787378	yogurt	0

In [92]:

user_order = pd.read_csv('orders.csv')[['user_id', 'order_id']]
user_order.head()

Out[92]:

	user_id	order_id
0	1	2539329
1	1	2398795
2	1	473747
3	1	2254736
4	1	431534

In [94]:

final = pd.merge(user_order, combined)[['user_id', 'aisle', 'reordered']]
print(final.shape)
final.head()

(33819106, 3)

Out[94]:

	user_id	aisle	reordered
0	1	soft drinks	0
1	1	popcorn jerky	0
2	1	soy lactosefree	0
3	1	paper goods	0
4	1	popcorn jerky	0

In [95]:

import numpy as np
%matplotlib inline
user_count = final.groupby('user_id').agg({'reordered' : [np.size, np.mean, np.std]})[:]
user_count.hist(('reordered', 'size'), bins=500)
print(user_count.shape)

(206209, 3)

In [102]:

user_count.loc[user_count[('reordered', 'size')] > 50, 'heavy'] = 1
grouped_idx = user_count[user_count['heavy'] == 1]
grouped_idx.hist(('reordered', 'size'), bins=500)
print(grouped_idx.shape)

(145776, 4)

In [103]:

final_filtered = final.loc[final['user_id'].isin(grouped_idx.index)]
final_filtered.head()

Out[103]:

	user_id	aisle	reordered
0	1	soft drinks	0
1	1	popcorn jerky	0
2	1	soy lactosefree	0
3	1	paper goods	0
4	1	popcorn jerky	0

In [104]:

print(final.shape)
print(final_filtered.shape)

(33819106, 3)
(32041336, 3)

In [105]:

pvt = final_filtered.pivot_table(index = 'user_id', columns = 'aisle', values = 'reordered')
pvt.head()

Out[105]:

aisle	air fresheners candles	asian foods	baby accessories	baby bath body care	baby food formula	bakery desserts	baking ingredients	baking supplies decor	beauty	beers coolers	...	spreads	tea	tofu meat alternatives	tortillas flat bread	trail mix snack mix	trash bags liners	vitamins supplements	water seltzer sparkling water	white wines	yogurt
user_id
1	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	0.0	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.500000
2	NaN	0.333333	NaN	NaN	NaN	NaN	0.000000	NaN	NaN	NaN	...	0.0	0.0	0.0	NaN	NaN	NaN	NaN	0.5	NaN	0.666667
3	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	...	0.5	0.0	NaN	NaN	NaN	NaN	NaN	0.5	NaN	NaN
7	NaN	NaN	NaN	NaN	NaN	NaN	0.333333	NaN	NaN	NaN	...	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	NaN	0.800000
8	NaN	0.000000	NaN	NaN	NaN	NaN	0.000000	NaN	NaN	NaN	...	NaN	NaN	0.0	0.0	NaN	NaN	NaN	NaN	NaN	NaN

5 rows × 134 columns

In [106]:

corr = pvt.corr()
corr.head()

Out[106]:

aisle	air fresheners candles	asian foods	baby accessories	baby bath body care	baby food formula	bakery desserts	baking ingredients	baking supplies decor	beauty	beers coolers	...	spreads	tea	tofu meat alternatives	tortillas flat bread	trail mix snack mix	trash bags liners	vitamins supplements	water seltzer sparkling water	white wines	yogurt
aisle
air fresheners candles	1.000000	0.091770	0.065998	-0.002700	0.077041	0.068956	0.095971	0.015324	0.113685	0.041045	...	0.086948	0.127405	0.049427	0.098696	0.054409	0.135516	0.102627	0.084800	0.132801	0.098967
asian foods	0.091770	1.000000	0.141231	0.122654	0.138802	0.118809	0.151334	0.117783	0.053513	0.111905	...	0.146299	0.123600	0.196432	0.153653	0.103492	0.136582	0.120203	0.118680	0.071340	0.156958
baby accessories	0.065998	0.141231	1.000000	0.055613	0.215751	0.117233	0.110193	0.079798	0.089913	0.058241	...	0.130325	0.116562	0.031429	0.157724	0.058691	0.178630	0.125893	0.171265	0.425205	0.192914
baby bath body care	-0.002700	0.122654	0.055613	1.000000	0.139511	0.009053	0.113821	0.066624	0.149574	0.108886	...	0.136989	0.090880	0.091536	0.111135	0.066639	0.147592	0.074446	0.104273	0.091513	0.156196
baby food formula	0.077041	0.138802	0.215751	0.139511	1.000000	0.126697	0.162811	0.072900	0.105111	0.219911	...	0.215741	0.129856	0.179175	0.171607	0.157447	0.181142	0.121044	0.192916	0.120679	0.290786

5 rows × 134 columns

In [107]:

corr.sort_values(by=('baby accessories'), ascending=False)['baby accessories']

Out[107]:

aisle
baby accessories        1.000000
white wines             0.425205
spirits                 0.326418
diapers wipes           0.241350
fresh fruits            0.226517
                          ...   
packaged produce        0.013749
energy sports drinks    0.008611
skin care              -0.013996
facial care            -0.052337
kosher foods           -0.057063
Name: baby accessories, Length: 134, dtype: float64

In [ ]:

In [ ]: