Big Data Multi-Strategy Predator Algorithm for Passenger Flow Prediction

Faced with the rapidly recovering tourism market, accurate prediction of passenger flow can help local authorities achieve more effective resource regulation. Therefore, based on big data technology, a multi-strategy predator algorithm is proposed, which uses the Marine Predator Algorithm, combined with regularized extreme learning machines and Collaborative Filtering Algorithms, to achieve accurate passenger flow prediction. The experiment findings denote that the performance parameters of the algorithm are excellent, with extremely strong convergence performance, and only 30 iterations are needed to reach the optimal solution. The fitting degree of this algorithm is 97.8%, which is 6.27% -19.31% higher than that of long and short-term memory networks, random forest algorithms, and support vector machine regression. In actual passenger flow prediction, the error rate of this algorithm is only 2.29%, which is 3.47% -6.50% higher than the three comparison algorithms. This study provides a new and efficient prediction method for passenger flow prediction. Its excellent predictive performance can not only help relevant departments predict and manage passenger traffic more accurately, but also provide reference for traffic prediction in other fields. Overall, this study has important reference value and practical significance for the research and practice of passenger flow prediction.

@article{Guo2024,
  title     = {Big Data Multi-Strategy Predator Algorithm for Passenger Flow Prediction},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {15},
  number    = {5},
  year      = {2024},
  publisher = {The Science and Information Organization},
  author    = {Peng Guo},
  doi       = {10.14569/IJACSA.2024.0150580},
  url       = {https://doi.org/10.14569/IJACSA.2024.0150580}
}