Improving hotel room demand forecasting with a hybrid GA-SVR methodology based on skewed data transformation, feature selection and parsimony tuning.
Rubén UrracaAndrés Sanz-GarcíaJulio Fernández-CenicerosAlpha V. Pernía-EspinozaFrancisco J. Martínez de Pisón AscacibarPublished in: Log. J. IGPL (2017)
Keyphrases
- demand forecasting
- hybrid ga
- feature selection
- skewed data
- support vector regression
- forecasting accuracy
- load balancing
- genetic algorithm ga
- historical data
- support vector machine
- support vector
- neural network
- class imbalance
- machine learning
- feature space
- text classification
- feature subset
- genetic algorithm
- regression model
- dimensionality reduction
- high dimensionality
- feature selection algorithms
- pattern recognition
- data sets
- model selection
- supply chain
- nearest neighbor