Problem Statement - Question 2
Future Group has built an attractive portfolio of some of the fastest growing consumer brands in India. Around 400 million customers walk into their stores each year and choose products and services supplied by over 30,000 small, medium and large entrepreneurs and manufacturers from across India.
Future Group employs a staggering 36000 people directly from every section of Indian society. Not just employment, the group wants to usher positive socio-economic changes across all section of India societies.
In retail, Big Bazaar is the most popular brand by Future Group. Big Bazaar has retail outlets across major metropolitan cities in India. The company wants to use machine learning to better understand customer behaviour and understand their buying needs better.
In this problem, you've to predict the products that a customer will buy in next 1 month.
Download Dataset
Data Description
The dataset consists of Big Bazar consumers. The data consists of customer purchases from May 2015 to June 2017.
BigBazaar runs various loyalty programs, festive offers which provide their customer more opportunities to avail discounts. Customers can use these offers or loyalty program to either avail discount or make payment.
products.csv
This file contains products transaction information and highlights the discounts used by customers in each transaction.
| Variable | Description |
|---|---|
| customerID | unique customer ID |
| DOB | date of birth of customer |
| Gender | gender |
| State | customer's state |
| PinCode | pincode of area where customer lives |
| transactionDate | date of transaction |
| store_code | unique code of big bazaar store |
| store_description | description of store |
| till_no | counter no. in the store |
| transaction_number_by_till | unique transaction number by counter, transactionDate, store_code |
| promo_code | if promotional code (offer) used in the transaction |
| promo_description | description of the offer |
| product_code | unique code of the product purchased |
| product_description | description of the product purchased |
| sale_price_after_promo | sale price of the product after applying promotion |
| discountUsed | after promo, customer used this discount(s) on transaction |
tenderModes.csv
This file contains information on payment mode(s) used by a customer in making a transaction.
| Variable | Description |
|---|---|
| customerID | unique customer ID |
| DOB | date of birth of customer |
| Gender | Gender |
| State | customer's state |
| PinCode | pincode of area where customer lives |
| transactionDate | date of transaction |
| store_code | unique code of big bazaar store |
| store_description | description of store |
| till_no | counter no. in the store |
| tender_type | mode used to make payment |
| transaction_number_by_till | unique transaction number by counter, transactionDate, store_code |
| payment_amount_by_tender | amount paid using the payment mode |
| PaymentUsed | description of mode of payment |
Submission Format
A participant has to submit a .csv file containing customerID against predicted products. Your submission file should have 39205 rows.
customerID, products
BBID_204221, '300663432,1000099534,1000475598,None,None,None,None...'
BBID_204254, '300663432,1000099534,1000475598,None,None,None,None...'
BBID_204830, '300663432,1000099534,1000475598,None,None,None,None...'
BBID_204880, '300663432,1000099534,1000475598,None,None,None,None...'
BBID_204910, '300663432,1000099534,1000475598,None,None,None,None...'
Evaluation Metric
Submission will be evaluated based on NDCG@k where k = 20 averaged across all users in the test set. Read more about NDCG here.
For every customer, you should provide a list of 20 products. In case there are no products, make prediction as "None".
Scripts
- Collaborative Filtering Recommender (Python) - Click Here
Note
- [Oct 17, 23:01] - Check Question 1 Here. You have to attempt both questions in order to proceed to next round.
