Spaces:
Sleeping
Sleeping
Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -1,13 +1,57 @@
|
|
| 1 |
-
---
|
| 2 |
-
title:
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
-
sdk: gradio
|
| 7 |
-
sdk_version:
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
title: Hotel Booking Cancellation Risk
|
| 3 |
+
emoji: 🏨
|
| 4 |
+
colorFrom: blue
|
| 5 |
+
colorTo: red
|
| 6 |
+
sdk: gradio
|
| 7 |
+
sdk_version: "4.44.0"
|
| 8 |
+
app_file: app.py
|
| 9 |
+
pinned: false
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# Hotel Booking Cancellation Risk Predictor
|
| 13 |
+
|
| 14 |
+
## Model description
|
| 15 |
+
|
| 16 |
+
A **HistGradientBoosting** classifier trained on the
|
| 17 |
+
[Hotel Booking Demand](https://www.kaggle.com/datasets/jessemostipak/hotel-booking-demand)
|
| 18 |
+
dataset (Antonio et al., 2019). Given eight booking parameters, the model returns the
|
| 19 |
+
probability that the booking will be cancelled.
|
| 20 |
+
|
| 21 |
+
**Test AUC:** 0.9067
|
| 22 |
+
|
| 23 |
+
## Intended use
|
| 24 |
+
|
| 25 |
+
Educational demo for the course *Introduction to Digital Content and Artificial Intelligence*,
|
| 26 |
+
Universitat de València (2025/2026). **Not intended for production use.**
|
| 27 |
+
|
| 28 |
+
## Inputs
|
| 29 |
+
|
| 30 |
+
| Field | Type | Description |
|
| 31 |
+
|---|---|---|
|
| 32 |
+
| Hotel type | Categorical | City Hotel or Resort Hotel |
|
| 33 |
+
| Lead time | Integer (days) | Days between booking and arrival |
|
| 34 |
+
| Deposit type | Categorical | No Deposit / Non Refund / Refundable |
|
| 35 |
+
| Market segment | Categorical | Channel through which booking was made |
|
| 36 |
+
| Special requests | Integer (0-5) | Number of special requests |
|
| 37 |
+
| Previous cancellations | Integer | Past cancellations by this guest |
|
| 38 |
+
| ADR | Float (€) | Average Daily Rate |
|
| 39 |
+
| Total nights | Integer | Length of stay |
|
| 40 |
+
|
| 41 |
+
## Output
|
| 42 |
+
|
| 43 |
+
Probability of cancellation, labelled as Low (< 30%), Medium (30–60%), or High (> 60%).
|
| 44 |
+
|
| 45 |
+
## Known limitations
|
| 46 |
+
|
| 47 |
+
- Trained on European hotel data (Portugal) from 2015–2017. May not generalise to other markets.
|
| 48 |
+
- Missing values for features not provided in the form are imputed with training-set medians.
|
| 49 |
+
- No concept drift detection — model performance may degrade over time.
|
| 50 |
+
- Labels are in English; the target audience may need localisation.
|
| 51 |
+
|
| 52 |
+
## Privacy & GDPR
|
| 53 |
+
|
| 54 |
+
This demo accepts no personally identifiable information (PII). No data entered in the
|
| 55 |
+
form is stored or logged. For a production deployment handling guest data, a full GDPR
|
| 56 |
+
impact assessment would be required and EU-resident cloud infrastructure should be used
|
| 57 |
+
(Azure West Europe or AWS eu-west regions).
|