Passenger flow forecast of tourist attraction based on MACBL in LBS big data environment

Abstract: The existing scenic spot passenger flow prediction models have poor prediction accuracy and inadequate feature extraction ability. To address these issues, a multi-attentional convolutional bidirectional long short-term memory (MACBL)-based method for predicting tourist flow in tourist scenic locations in a location-based services big data environment is proposed in this study. First, a convolutional neural network is employed to identify local features and reduce the dimension of the input data. Then, a bidirectional long short-term memory network is utilized to extract time-series information. Second, the multi-head attention mechanism is employed to parallelize the input data and assign weights to the feature data, which deepens the extraction of important feature information. Next, the dropout layer is used to avoid the overfitting of the model. Finally, three layers of the above network are stacked to form a deep conformity network and output the passenger flow prediction sequence. In contrast to the state-of-the-art models, the MACBL model has enhanced the root mean square error index by at least 2.049, 2.926, and 1.338 for prediction steps of 24, 32, and 60 h, respectively. Moreover, it has also enhanced the mean absolute error index by at least 1.352, 1.489, and 0.938, and the mean absolute percentage error index by at least 0.0447, 0.0345, and 0.0379% for the same prediction steps. The experimental results indicate that the MACBL is better than the existing models in evaluating indexes of different granularities, and it is effective in enhancing the forecasting precision of tourist attractions.

Object Details

• Location: Deutsche Nationalbibliothek Frankfurt am Main
• Extent: Online-Ressource
• Language: Englisch

References and Relationships

• Bibliographic citation: Passenger flow forecast of tourist attraction based on MACBL in LBS big data environment ; volume:15 ; number:1 ; year:2023 ; extent:12 ; Open Geosciences ; 15, Heft 1 (2023) (gesamt 12)

Contributors, Places and Time

• Creator: Tang, Qili ; Yang, Li ; Pan, Li

Further information

• DOI: 10.1515/geo-2022-0577
• URN: urn:nbn:de:101:1-2023122013050927077541
• Rights: Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
• Last update: 15.08.2025, 7:22 AM CEST