Bank Marketing

Problem Description

Description

Amid uncertainties related to inflation, a banking institution is conducting a telemarketing campaign to convince customers to open term deposits. After a long promotion program, the company wants to evaluate the effectiveness of the campaign. The dataset contains information about customers, demographic characteristics, and details about the calls made during the marketing campaign. The goal of the problem is to analyze this data and solve several subtasks related to customer behavior.

Input Data

Participants receive three files:

train.csv — The training dataset. It contains all variables, including the target variable deposit.
test.csv — The testing dataset. It contains the same variables as train.csv, but without the deposit column.
sample_output.csv — A file whose structure resembles the submission that must be sent. This receives 0 points.

Column Description

Column	Description
`id`	the id
`age`	customer's age
`job`	customer's occupation
`marital`	marital status
`education`	educational level
`default`	whether the customer has ever defaulted
`balance`	account balance
`housing`	whether the customer has a housing loan
`loan`	whether the customer has a personal loan
`contact`	contact type
`day`	day of the call
`month`	month of the call
`duration`	call duration
`campaign`	number of calls during the campaign
`pdays`	number of days since the last campaign
`previous`	number of previous contacts
`poutcome`	outcome of the previous campaign
`deposit`	target variable: whether the customer opened a deposit

Requirements

The problem is divided into four subtasks.

Subtask 1 (10 points)

Determine the occupation (job) with the highest deposit rate.

The deposit rate represents the proportion of customers for whom deposit = 1 out of the total customers with that job.

Output: the occupation.

Subtask 2 (15 points)

Determine the month (month) in which the highest number of customers from the training dataset were contacted by phone.

Output: the name of the month.

Subtask 3 (55 points)

Build a binary classification model that predicts the deposit variable.

The model must be trained using train.csv and must generate predictions for test.csv.

Subtask 4 (20 points)

Group the customers from test.csv into two clusters based on numerical variables and return the cluster label for each customer.

Output File Format

The solution must generate a submission.csv file with the following structure:

Python
1subtaskID,datapointID,answer21,1,management32,1,jan43,5686,053,5206,16...74,5686,184,5206,09...

Explanations:

For Subtasks 1 and 2, the datapointID column will have the value 1.
For Subtasks 3 and 4, the datapointID column will have the id value from test.csv.

Evaluation

Subtask 1 — Exact answer verification.

Subtask 2 — Exact answer verification.

Subtask 3 — Classification is evaluated using F1 macro:

F1 macro	Score
>= 0.85	maximum score (55p)
< 0.65	0 points
between values	proportional score

Subtask 4 — Clustering is evaluated using Adjusted Rand Index (ARI):

ARI	Score
>= 0.9	maximum score (20p)
< 0.5	0 points
between values	proportional score

Files

Download problem files

Submit Solution

Upload the result file manually. You can also attach source code.100 MB maximum in total

Submission File

Click to upload or drag and drop

CSV, ZIP, etc. (MAX. 100MB)

Source Code File (optional)

Click to upload or drag and drop

Archive, notebook or code file

Work and submit from Google Colab

Open the notebook prepared for this problem. The first cell downloads the data; when you are done, submit your solution for evaluation directly from the notebook.

Colab submissions use the same evaluator and limits. Track the result in My Submissions.

MLCompete can access only the notebook it creates in Drive, not your other files.

Doesn’t open? Use a temporary code

Generate the code here, open the public notebook, and enter it when the first cell asks.

Open the public notebook in Colab

Paste the workspace code only when the notebook asks for it. Never share it with someone else.