Design and Implementation of a Personalized Tourism Recommendation System Based on the Data Mining and Collaborative Filtering Algorithm

• Published in: Computational Intelligence and Neuroscience Vol. 2022; pp. 1 - 14
• Main Authors: Nan, Xiang, Kayo kanato, Wang, Xiaolan
• Format: Journal Article
• Language: English
• Published: United States Hindawi 31.08.2022; Hindawi Limited; John Wiley & Sons, Inc
• Subjects: Accuracy; Algorithms; Banks (Finance); Collaboration; Context; Customization; Data analysis; Data Mining; Data processing; Efficiency; Filtration; Humans; Information management; Knowledge Bases; Knowledge bases (artificial intelligence); Knowledge management; Markov analysis; Ontology; Preferences; Recommender systems; Research Article; Route optimization; Tourism; Transfer learning; Travel; Travel industry; User generated content
• ISSN: 1687-5265; 1687-5273; 1687-5273
• DOI: 10.1155/2022/1424097

A personalized tourism recommendation system provides convenient and economically affordable travel information for individuals/groups. This recommendation system banks on accumulated and analyzed data for providing context-aware travel solutions. For improving the recommendation efficiency and data analysis of such systems, this article introduces a mining and filtering harmonized collaborative process, named as the collaborative mining and filtering process (CMFP), for reducing the data processing overheads and improving the recommendation ratio. In this process, the accumulated data from the global and personal travel, expenditure, and other information are collaboratively analyzed. This analysis is powered by knowledge-based transfer learning for reducing the retardation in the large data processing. Based on the context-based data analysis, the filtering and mining are jointly performed for providing recommendations. In the filtering process, the maximum processed contextual data are extracted for updating the current knowledge base. From this base, the recommendation for adaptable travel is recommended for the user. This process’s performance is analyzed using the metrics accuracy, data handling rate, mining time, and overhead.