greygroup_​corporatedata

Random Forest

Data Cleaning

Decision Tree

Logistic Regression

Naive Bayes

Data Understanding

Dataset Overview

• Dataset: Hotel Booking Demand

• Objective: Predict whether a hotel booking will be canceled.

• Target Variable: is_canceled

  • 0 = Not Cancelled

  • 1 = Cancelled

Dataset Characteristics

• Dataset type - Hotel booking records.

• Prediction problem - Hotel booking cancellation prediction

• Hotel types - City hotel & resort hotel

• Data types - Numerical & categorical

• Number of records - 119,390 hotel bookings

• Target variable - is_canceled

Dataset Characteristics

• Customer characteristics - adults, children, babies, customer_type

• Booking Informatio - nlead_time, booking_changes, previous_cancellations

• Reservation Details - hotel, meal, reserved_room_type, assigned_room_type

• Marketing Information - market_segment, distribution_channel

• Financial Information - adr, deposit_type

• Historical Behaviour - previous_bookings_not_canceled

• Target Variable - is_canceled

Data Exploration

• The dataset contains a mixture of numerical and categorical variables.

• The response variable (is_canceled) is binary, making the dataset suitable for classification algorithms.

• Numerical variables such as lead_time, adr, and days_in_waiting_list exhibit wide value distributions.

• Several categorical variables contain multiple categories that require transformation before modelling.

• Some variables potentially have strong relationships with booking cancellations, particularly lead_time, deposit_type, market_segment, and previous_cancellations.

Data Quality Assessment

• agent - Missing values

• company - Large number of missing values

• country - Missing values

• children - Small number of missing values

Data Types

• Numerical variables

• Nominal variables

• Binary variables

Dashboard Component