一、简介 ELO商户类别推荐有助于了解客户忠诚度

Elo Merchant Category Recommendation Help understand customer loyalty (ELO商户类别推荐有助于了解客户忠诚度)

竞赛描述：

想象一下，在一个陌生的地方饿着肚子，在适当的时候，根据你的个人喜好，提供餐馆推荐。推荐信附带了您的信用卡提供商提供的拐角处当地饭店的折扣！

目前，作为巴西最大的支付品牌之一，ELO已经与商家建立了合作关系，为持卡人提供促销或折扣。但是，这些促销活动对消费者或商家都有效吗？客户喜欢他们的体验吗？商人看到重复的生意了吗？个性化是关键。

ELO建立了机器学习模型，以了解客户生命周期中从食品到购物的最重要方面和偏好。但到目前为止，没有一个是专门为个人或个人资料定制的。这就是你进来的地方。

在这场竞争中，Kaggers将通过发现客户忠诚度中的信号，开发算法来识别并为个人提供最相关的机会。您的输入将改善客户的生活，帮助ELO减少不必要的活动，为客户创造正确的体验。

评价指标

提交的数据按均方根误差（root mean squared error）计分。RMSE定义为：

where $\hat{y}$

提交文件格式：

card_id, target
C_ID_9e86007114,0
C_ID_1c9f77086c,0.5
C_ID_07b20e9908,0
C_ID_63d6bac69a,0
C_ID_bbc26a86eb,0
C_ID_f749aad790,0
C_ID_7b5c15ff41,-0.25
C_ID_ec6b0f2d30,0
C_ID_0a11e759c5,0

Timeline　　

February 19, 2019 - External Data Disclosure deadline. All external data used in the competition must be disclosed in the forums by this date.

数据

需要的文件

train.csv and test.csv 文件包含用于训练和预测的card_ids
The historical_transactions.csv and new_merchant_transactions.csv files 包含有关每张卡交易的信息。
historical_transactions.csv 包含在任何提供的商户ID上的每张卡最多3个月的交易价值。
new_merchant_transactions.csv 包含两个月内新商户的交易（该卡尚未访问的商户ID）。
merchants.csv 包含数据集中表示的每个商户ID的汇总信息。

The historical_transactions.csv and new_merchant_transactions.csv files contain information about each card's transactions.
historical_transactions.csv contains up to 3 months' worth of transactions for every card at any of the provided merchant_ids.
new_merchant.csv contains the transactions at new merchants (merchant_ids that this particular card_id has not yet visited) over a period of two months.
merchants.csv contains aggregate information for each merchant_id represented in the data set.

数据格式如下：

train.csv和test.csv包含卡ID和卡本身的信息-卡激活的第一个月（ the first month），等等。train.csv也包含目标。
historical_transactions.csv and new_merchant_transactions.csv are designed to be joined with train.csv, test.csv, and merchants.csv.它们包含有关每张卡的交易信息。
merchants can be joined with the transaction sets to provide additional merchant-level information. 可以将商家与交易集结合起来，以提供额外的商家级别信息。

预测：

You are predicting a loyalty score for each card_id represented in test.csv and sample_submission.csv.

文件描述：

train.csv - the training set
test.csv - the test set
sample_submission.csv - a sample submission file in the correct format - contains all card_ids you are expected to predict for.
historical_transactions.csv - up to 3 months' worth of historical transactions for each card_id
merchants.csv - additional information about all merchants / merchant_ids in the dataset.
new_merchant_transactions.csv - two months' worth of data for each card_id containing ALL purchases that card_id made at merchant_ids that were not visited in the historical data.