Application of genetic algorithms to optimize a truncated mean k-nearest neighbours regressor for hotel reservation forecasting

Authors:
Andrés Sanz-García;Julio Fernández-Ceniceros;Fernando Antoñanzas-Torres;F. Javier Martínez-de-Pisón-Ascacibar
Affiliations:
EDMANS Research Group, University of La Rioja, Logroño, Spain;EDMANS Research Group, University of La Rioja, Logroño, Spain;EDMANS Research Group, University of La Rioja, Logroño, Spain;EDMANS Research Group, University of La Rioja, Logroño, Spain
Venue:
HAIS'12 Proceedings of the 7th international conference on Hybrid Artificial Intelligent Systems - Volume Part I
Year:
2012

Citing 8
Cited 0

Instance-Based Learning Algorithms

Machine Learning
Forcasting uncertain hotel room demand

Information Sciences: an International Journal
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
An Introduction to Genetic Algorithms

An Introduction to Genetic Algorithms
Editorial: Hybrid intelligent algorithms and applications

Information Sciences: an International Journal
A soft computing method for detecting lifetime building thermal insulation failures

Integrated Computer-Aided Engineering
Neural visualization of network traffic data for intrusion detection

Applied Soft Computing
Editorial: Editorial: New trends and applications on hybrid artificial intelligence systems

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Progress in information technologies and their broad social expansion have led to the appearance of websites specialized in online hotel booking. These sites offer new approaches for customers that have demonstrated a strong tendency towards making last minute reservations. This scenario has dramatically affected the task of predicting hotel bookings, making these estimations is now much more complex using the traditional forecasting models. Given the importance of this estimation, it is crucial to find more accurate prediction models that take into account these new situations. This work aims to develop an application to predict hotel room reservations that tackles the consequences of last-minute reservations. Our proposal combines genetic algorithms optimization and truncated mean k-nearest neighbours regressor. After the analysis, we conclude that our method shows a significant improvement regarding working with historical booking information when compared with classical models. Therefore, the results of the study illustrate that our application will enable the development of useful prediction demand calendars.