基于大数据的卷烟品牌转换行为研究——以Y市为例
Research on Cigarette Brand Conversion Behavior Based on Big Data—Taking Y City as an Example

在大数据的时代背景下，为充分挖掘卷烟消费数据价值，本文以卷烟品牌转换行为为切入点，研判当前卷烟市场消费趋势，在此基础上寻找品牌替代关系。基于聚合支付收集到的海量消费数据，通过构建卷烟品牌转换行为的识别方法，分析和测算消费结构升级、降级和无明显变化的消费人群特征及占比，以此作为研判卷烟市场消费趋势的依据。并结合不同消费人群的交易行为特征，运用Logistic回归模型，识别卷烟品牌潜在转换行为。同时，追踪已经发生卷烟品牌转换行为的消费者购买品规的前后变化，借助共词分析方法，寻找不同卷烟品规之间的替代关系。该方法能有效反应卷烟市场消费趋势和品牌替代关系，为制定卷烟营销策略提供参考。
In the context of the era of Big data, in order to fully tap the value of cigarette consumption data, this paper takes cigarette brand conversion behavior as the starting point, studies and judges the current cigarette market consumption trend, and on this basis, looks for brand substitution relations. Based on the massive consumption data collected through aggregated payment, a recognition method for cigarette brand transformation behavior is constructed to analyze and calculate the characteristics and proportion of consumer groups with upgraded, downgraded, and no significant changes in consumption structure, as a basis for studying and judging the consumption trend of the cigarette market. Combining the transaction characteristics of consumers with different switching behaviors, a logistic regression model is used to identify potential switching behaviors of cigarette brands. At the same time, track the changes in the purchase specifications of consumers who have undergone cigarette brand switching behavior, and use co word analysis method to find alternative relationships between different cigarette specifications. This method can effectively reflect the consumption trend and brand substitution relationship in the cigarette market, providing reference for formulating cigarette marketing strategies.