基于决策树方法的欠发达地区城市居民旅游花费偏好研究——以江西赣州、吉安为例

A Decision Tree Analysis of Tourism Expenditure Preference of Urban Residents in Less-developed Areas: The Case of Ganzhou and Ji' an

以江西赣州和吉安两地为例,研究欠发达地区城市居民的旅游花费预算及其影响因素。首先,通过线性回归分析初步筛选了影响旅游花费预算的相关因素。然后,通过决策树分析方法,探明旅游花费预算的影响因素构成,证实了旅游天数、住宿设施类型选择、月收入水平是旅游花费预算的最有效预测变量,且这四者之间存在一定的非线性关系,由此细分出5类花费特征显著的潜在消费者群体。

Based on the survey in Ganzhou and Ji＇an of Jiangxi province, this paper carries out a research on travel expenditure budget and its influencing factors of urban residents in less-developed areas. Firstly, the related influencing factors of travel expenditure budget is preliminary selected through linear regression analysis; then decision tree analysis is adopted to ascertain the real composition of the influencing factors. The results show that the days of travelling, types of accommodation and monthly income are the most effective predictive variables, and there is some non-linear relation- ship among them. Finally, five segments of consumer markets with distinctive expenditure characteristics are subdivided according to the results.