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-CenicerosEnrique Sodupe-OrtegaFrancisco J. Martínez de Pisón AscacibarPublished in: HAIS (2015)
Keyphrases
- demand forecasting
- hybrid ga
- feature selection
- skewed data
- support vector regression
- genetic algorithm ga
- support vector machine
- load balancing
- forecasting accuracy
- historical data
- support vector
- class imbalance
- neural network
- regression model
- feature set
- feature extraction
- data mining
- dimensionality reduction
- supply chain
- model selection
- text classification
- feature subset
- classification accuracy
- feature space
- high dimensionality
- feature selection algorithms
- knn
- streaming data