Improving migration forecasting for transitory foreign tourists using an Ensemble DNN-LSTM model

The tourism industry is a key component in many nations' economies, and precise Forecasting of the number of arrivals of transient international visitors is necessary for efficient resource allocation and strategic planning. Traditional approaches for predicting time series need to catch up when capturing the intricate patterns and interdependencies in data about visitor movement. To overcome this obstacle, we have developed a model that combines Ensemble Deep Neural Networks (EDNN) with Long Short-Term Memories (LSTM) to enhance migration predictions. EDNNs learn complex patterns from historical data, whereas LSTMs model sequential relationships. We start with data from the study-based Migration Forecasting for Transitory Foreign Tourists. The data was preprocessed using Adaptive mean filter (AMF) methods. We used Linear Discriminant Analysis (LDA) to look at different configurations for extracting features. The EDNN-LSTM method had better results than its comparison in terms of a wide range of performance measures having an accuracy of 97%, an MAE of 40.75%, an MSE of 33.10%, and an RMSE of 30.10%. According to the results, the EDNN-LSTM model can represent the intricate dynamics of transitory foreign visitor movement. Consequently, this model is a useful resource for tourism stakeholders, policymakers, and enterprises.