Analysing Chinese citizens' intentions of outbound travel: a machine learning approach

Due to the tremendous expansion of Chinese outbound travel market, it is of great significance to identify the characteristics of potential Chinese outbound tourists so as to develop targeted marketing strategies. As both extrinsic and intrinsic characteristics of individuals are important in influencing their decision-making process, this study explores potential Chinese outbound tourists from these two aspects. Demographic and psychographic factors, as the proxy of extrinsic and intrinsic characteristics, respectively, are incorporated into this study to construct the relationship between tourists' characteristics and their intentions of outbound travel. An advanced machine learning approach, called twice-learning, is employed for modelling in this study. As an intelligent data analysis tool, this method is able to construct models that can provide insight into the ground-truth relationships hidden beneath the data in an essentially comprehensible way, without being limited by the typical assumptions held by the traditional data analysis methods. By applying this method, the important personal factors that influence Chinese citizens' intentions of outbound travel are detected, and the typical groups of potential Chinese outbound tourists are characterised. The findings would be beneficial for destination marketers to develop marketing strategies on positioning and advertising which are tailored to potential Chinese outbound travel market.