Effects of recommender systems different information based on tourism behaviour intention.

Abstract

Tourism recommender systems include multiple information tools to foster booking decisions. Based on an experimental methodology, this research objective is to determine if there exist differences in the determinants of behavioural intention. Two randomly selected samples of university students were collected for each scenario designed, one using an artificial intelligence-based recommender system and another using a traditional information-based recommender system. The choices were in all cases unknown brands, and each sample unit had the following similar characteristics to avoid bias: frequent online tourism agencies' usage; travelling at least once during the last year; the accommodation chosen for this research should not be familiar to them. The study results confirmed a positive effect of the determinants studied in both recommender systems analysed. In addition, some differences were discovered. For the case of an artificial intelligence-based recommender system, the effect of perceived quality on satisfaction is double, and the effect of satisfaction in behaviour intention is also higher. These results highlight which antecedents of the tourist booking decision-making process are more deeply impacted by including artificial intelligence-based information of choices in recommender systems. Moreover, our results may help hoteliers and marketers understand the possible effects of artificial intelligence inclusion in recommender systems on tourist decision making and behaviour intention.