fetch_telco_customer_churn#

empulse.datasets.fetch_telco_customer_churn(*, backend, data_home=None, download_if_missing=True)[source]#

Fetch the Telco Customer Churn dataset from OpenML (binary classification).

The goal is to predict whether a customer will churn or not. The target variable is whether the customer churned, 1 = churned, 0 = active.

For additional information about the dataset, consult the User Guide.

Classes

2

Churners

1869

Non-churners

5163

Samples

7032

Features

19
Parameters:
backendmodule

Dataframe library to use for data and target. Pass the library module directly, e.g. backend=polars or backend=pandas.

data_homestr or Path, optional

Directory used for caching downloaded data. Defaults to ~/empulse_data (or $EMPULSE_DATA_HOME).

download_if_missingbool, default=True

If False, raise an OSError when the data is not cached locally.

Returns:
datasetDataset

instance_costs contains {'monthly_charges': array} — the customer’s monthly bill amount, which scales the false-negative and false-positive costs.

Notes

Cost matrix (Petrides & Verbeke 2021, Vanderschueren et al. 2022):

Actual positive \(y_i = 1\)

Actual negative \(y_i = 0\)

Predicted positive \(\hat{y}_i = 1\)

tp_cost \(= 0\)

fp_cost \(= fp\_months \cdot monthly\_charges_i\)

Predicted negative \(\hat{y}_i = 0\)

fn_cost \(= fn\_months \cdot monthly\_charges_i\)

tn_cost \(= 0\)

The cost matrix uses symbolic parameters with the following defaults:

  • fn_months = 12.0 (annual revenue lost upon churn)

  • fp_months = 2.0 (two months of retention offer cost)

References

[1]

Petrides, G., & Verbeke, W. (2021). Cost-sensitive ensemble learning: a unifying framework. Data Mining and Knowledge Discovery, 1–28.

[2]

Vanderschueren, T., Verdonck, T., Baesens, B., & Verbeke, W. (2022). Predict-then-optimize or predict-and-optimize? An empirical evaluation of cost-sensitive learning strategies. Information Sciences, 594, 400–415.

Examples

import numpy as np
import pandas as pd
from empulse.datasets import fetch_telco_customer_churn
from empulse.metrics import Metric, Cost

dataset = fetch_telco_customer_churn(backend=pd)

# replace with your own model's predicted probabilities
y_score = np.random.default_rng(0).uniform(size=len(dataset.target))

metric = Metric(dataset.cost_matrix, Cost())
score = metric(dataset.target, y_score, **dataset.instance_costs)