Forecasting Tourist Arrivals via Random Forest and Long Short-term Memory

Forecasting resort hotel tourism demand using deep learning techniques - A systematic literature review

In the hospitality industry, revenue management is vital for the sustainability of the business. Powering this strategic concept is the occupancy rate (OR) forecast. Predicting occupancy of the hotel is essential for managers' decision-making process as it gives an estimate of the future business performance. However, the fast-changing marketing demands in the tourism sector, boosted by the advent of online booking, generating accurate forecast figures is nowadays a tough task - needing personnel with advance technical skills and expensive software. The aim of the Systematic Literature review is to provide an insight of the use of Deep Learning techniques for OR prediction. The latest trends in this field over five years (from 2017 to 2022) are highlighted. Through this SRL, three research questions are answered. The questions are related to the variables, deep learning algorithms for prediction and the evaluation metrics used for evaluating the models developed. The Snowballing methodology was used to carry out the SLR. 50 papers were selected for the final analysis. Five categories of variables were identified. LSTM was found to be the most popular deep learning algorithm used to build prediction models. Seven performance metrics were found and among them MAPE was the most popular. To conclude it was found that the hybrid model, CNN-LSTM, to increase accuracy and required more investigation.

PM2.5 Concentration Prediction Based on Spatiotemporal Feature Selection Using XGBoost-MSCNN-GA-LSTM

With the rapid development of China’s industrialization, air pollution is becoming more and more serious. Predicting air quality is essential for identifying further preventive measures to avoid negative impacts. The existing prediction of atmospheric pollutant concentration ignores the problem of feature redundancy and spatio-temporal characteristics; the accuracy of the model is not high, the mobility of it is not strong. Therefore, firstly, extreme gradient lifting (XGBoost) is applied to extract features from PM2.5, then one-dimensional multi-scale convolution kernel (MSCNN) is used to extract local temporal and spatial feature relations from air quality data, and linear splicing and fusion is carried out to obtain the spatio-temporal feature relationship of multi-features. Finally, XGBoost and MSCNN combine the advantages of LSTM in dealing with time series. Genetic algorithm (GA) is applied to optimize the parameter set of long-term and short-term memory network (LSTM) network. The spatio-temporal relationship of multi-features is input into LSTM network, and then the long-term feature dependence of multi-feature selection is output to predict PM2.5 concentration. A XGBoost-MSCGL of PM2.5 concentration prediction model based on spatio-temporal feature selection is established. The data set comes from the hourly concentration data of six kinds of atmospheric pollutants and meteorological data in Fen-Wei Plain in 2020. To verify the effectiveness of the model, the XGBoost-MSCGL model is compared with the benchmark models such as multilayer perceptron (MLP), CNN, LSTM, XGBoost, CNN-LSTM with before and after using XGBoost feature selection. According to the forecast results of 12 cities, compared with the single model, the root mean square error (RMSE) decreased by about 39.07%, the average MAE decreased by about 42.18%, the average MAE decreased by about 49.33%, but R2 increased by 23.7%. Compared with the model after feature selection, the root mean square error (RMSE) decreased by an average of about 15%. On average, the MAPE decreased by 16%, the MAE decreased by 21%, and R2 increased by 2.6%. The experimental results show that the XGBoost-MSCGL prediction model offer a more comprehensive understanding, runs deeper levels, guarantees a higher prediction accuracy, and ensures a better generalization ability in the prediction of PM2.5 concentration.