Personalized travel recommendation: a hybrid method with collaborative filtering and social network analysis

ABSTRACT

This study proposes a hybrid method for producing personalized travel recommendation that better meet travellers’individual needs and also improve their online booking experience. The proposed method integrates multi-attribute collaborative filtering with social network analysis within the framework of large-scale group decision-making. It includes four modules, i.e. identification of online opinion experts, construction of a social network, detection of user communities, and interactively produced of personalized travel recommendation. Specifically, the preliminary user filtering and k-means clustering approach are utilized to identify the online opinion experts for a specific travel recommendation issue. Then, social network construction and its community detection process are adopted to alleviate the sparsity problem. Finally, the travel alternatives are ranked to select recommendations, and this is done interactively with travellers to handle the cold start problem. With the proposed method, a better online booking experience can be achieved for travellers, as they are presented with a more appropriate set of recommended options and so can make better travel decisions.

KEYWORDS:

• Personalized travel recommendation
• group decision-making
• multi-attribute collaborative filtering
• social trust network
• user community detection

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by National Natural Science Foundation of China: [grant number 71971124, 71932005]; the Liberal Arts Development Fund of Nankai University: [grant number ZB21BZ0106]; the China Postdoctoral Science Foundation: [grant number 2019M661000]; the Fundamental Research Funds for the Central Universities: [grant number 63202074]; the Humanities and Social Science Fund of Ministry of Education of China: [grant number 20YJC630002].