摘要
文章从需求频次与消费意愿的影响因素入手,提出了茶叶消费者粘性指数综合评价模型,该模型包括粘性构成要素、影响因素及其具体评价3部分;其中粘性构成要素包括需求频次与消费意愿,影响因素包括感知有用性、可替代性、感知价值、转移成本、专业素养和社会属性,具体评价指标共12项。采用调查问卷收集不同消费者对模型指标的评价数据,并通过了结构效度检验。采用随机森林算法分析各因素对茶叶消费者粘性的影响,以根据构成要素计算得出的粘性数值作为样本标签,以不同影响因素的具体评价指标作为样本特征,训练模型并进行参数调优,得到的最优模型在测试集上的MSE为481.36,模型拟合较好。引入SHAP值算法计算出不同影响因素的重要性程度,发现消费意愿的影响因素中,转移成本与社会属性重要性最高,权重值分别为0.343、0.325;需求频次的影响因素中,感知有用性重要性程度为0.184,高于可替换性的重要性。对不同影响因素进行分析,在此基础上提出关于茶产品的改进建议,进一步推动茶产品的改善与提升,促进消费者的茶产品消费,从而带动贵州省茶产业的深入发展。
In order to analyze the influencing factors of tea consumer stickiness and explore the influence degree of different factors on consumer stickiness,starting from the influencing factors of demand frequency and consumption intention,this paper put forward a comprehensive evaluation model of tea consumer stickiness,which included three parts:stickiness components,influencing factors and specific evaluation.The stickiness components included demand frequency and consumption intention.The influencing factors included perceived usefulness,substitutability,perceived value,transfer cost,professional quality and social attributes.There were 12 specific evaluation indicators.The questionnaire was used to collect the evaluation data of different consumers,and passed the structural validity test.The Random Forest Algorithm was used to analyze the influence of the factors.The stickiness value was used as the sample label,and the specific evaluation indexes were used as the sample characteristics.The model was trained and the parameters were optimized.The MSE of the optimal model on the test set was 481.36,and the model was well fitted.It was found that among the influencing factors of consumption intention,transfer cost and social attribute were the most important,and the weight values were 0.343 and 0.325 respectively.Among the influencing factors of demand frequency,the importance of perceived usefulness was 0.184,which was higher than that of substitutability.Based on the analysis,this paper put forward suggestions on the improvement of tea products,further promoted the improvement of tea products,and promoted the consumption of tea products,so as to drive the in-depth development of tea industry in Guizhou Province.
作者
付饶
王书博
刘智权
李雪
方茂达
FU Rao;WANG Shubo;LIU Zhiquan;LI Xue;FANG Maoda(School of Mathematics and Statistics,Guizhou University,Guiyang 550025,China)
出处
《中国茶叶》
2022年第1期37-43,共7页
China Tea
关键词
消费者粘性
综合评价模型
随机森林算法
SHAP值
重要性评价
consumer stickiness
comprehensive evaluation model
random forest algorithm
SHAP value
importance evaluation
