Research on Multidimensional Tourism Suitability Evaluation Model Based on Scenic Area Information and Time Series Data

In the context of the rapid development of global tourism, the evaluation of tourism suitability has become an important means to optimize the management and operation of scenic spots. This study aims to improve the prediction accuracy of tourism suitability and provide accurate and effective decision basis for scenic spots. Based on scenic spot information and time series data, this study combines long and short term memory network and dynamic time distortion algorithm to achieve multi-dimensional tourism suitability prediction. The results show that the LSTM-DTW model has the best performance in single dimension prediction, and the optimized model has the best performance in multi-model multidimensional prediction, with RMSE reaching 1.832, MAE reaching 1.362, R2 reaching 0.938, and MAPE reaching 9.92%. It can be seen that the combination of dynamic time distortion and deep learning helps to improve the prediction accuracy of tourism suitability, and provides a new research Angle and method for optimizing scenic spot management and improving tourism service quality.