论文部分内容阅读
用户在线机票代理商选择行为的分析对旅游搜索网站机票搜索产品系统的改进非常重要。该文基于某旅游搜索网站机票搜索和预订用户的在线数据,对该网站的用户在线代理商选择行为进行分析。运用了多项logit模型理论及方法,将有关代理商的机票报价、评价、增值服务、页面位置以及市场份额的变量引入模型,并建立了可以解释该选择行为的对数价格多项logit模型。结果显示:用户的确最关注的是代理商的机票报价;搭售保险和评价人数过少都不利于用户对代理商的选择;增值服务的提供和较高的评分会增加用户对代理商的兴趣;用户更倾向于选择展示页面中靠前位置的代理商,其次是靠后的代理商;市场份额大的代理商对用户的选择行为会产生显著影响。预测检验表明该文建立的模型比历史数据预测方法有更高的预测精度。
The analysis of the user’s online ticket agent selection behavior is very important to the improvement of the flight search product system of the travel search website. Based on the online data of a travel search website ticket search and booking user, this article analyzes the online agent selection behavior of users of the website. A number of logit model theories and methods are used to introduce the models of agents’ ticket pricing, evaluation, value-added services, page location and market share variables. Logistic models of logarithmic prices that can explain the selection behavior are also established. The results show that: the user is indeed the most concerned about the agent’s ticket price; tying insurance and evaluation of the number of people are not conducive to the choice of agents; value-added services and higher ratings will increase the user’s interest in agents ; Users are more inclined to choose to display the front page of the agents in the page, followed by the agents; market share of the agents of the user’s choice behavior will have a significant impact. Prediction tests show that the model established in this paper has higher prediction accuracy than the historical data forecasting method.