Title: Tourism data mining and analysis methods for smart tourism

Authors: Jiajun Chen; Guanxi Chen

Addresses: School of Business, Nantong Institute of Technology, Nantong, Jiangsu, China ' School of Design, NingboTech University, Ningbo, Zhejiang, China

Abstract: Smart tourism has shortcomings in real-time and information processing efficiency, and traditional data processing makes it difficult to cope with the rapid flow and complex analysis requirements of large-scale real-time data. This paper uses a distributed streaming data mining method based on Apache Flink. First, Flink is used to integrate scenic area sensors and multi-source data streams; then, the Long-Short-Term Memory (LSTM) algorithm is used to predict traffic trends; next, streaming K-means is used to mine tourist behaviour patterns; finally, dynamic optimisation schemes are generated through Frequent Pattern-Stream (FP-Stream). Studies have shown that this method achieves a low latency of 0.38 seconds and a high throughput of 5600 records/second when processing millions of data in a 20-node cluster through efficient parallel processing of distributed architecture and real-time analysis of streaming algorithms, providing real-time and precise technical support for smart tourism.

Keywords: smart tourism; distributed stream data mining; traffic trend prediction; tourist behaviour patterns; dynamic optimisation schemes.

DOI: 10.1504/IJCAT.2025.149867

International Journal of Computer Applications in Technology, 2025 Vol.77 No.1/2, pp.146 - 158

Received: 13 Mar 2025
Accepted: 26 Jun 2025
Published online: 14 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article