电子商务环境下商品的个性化定价研究

发布时间：2018-04-18 11:30

本文选题：个性化定价 + 支付意愿　；参考：《华中师范大学》2014年硕士论文

【摘要】：在消费者需求日期差异化的背景之下,我们进入了个性化时代,通过对消费者个人信息的分析研究,提供更加符合消费者喜好的产品及服务,尤其在电子商务环境下,这种个性化的服务更加方便快捷,也更加受到消费者的青睐。一方面,电子商务的个性化产品、个性化服务、个性化推荐等迅速发展,但是目前研究者对电子商务个性化的营销研究,大都倾向于研究个性化的产品和服务的推荐,而很少涉及个性化的价格和促销方式,使得个性化定价没有同期发展、受人瞩目。另一方面,近年来,随着人们对个性化的追求、电子商务个性化服务的发展,个性化定价作为个性化服务的重要组成部分,越来越受到人们的关注,加上数据挖掘技术的日臻成熟,业内对个性化定价在电子商务环境下的应用前景持乐观态度。本文从消费者个人行为、电子商务产品定价环境等方面入手,分析了个性化定价的主要步骤。针对个性化定价主要的三个步骤：测量支付意愿、搜索目标顾客、实施个性化定价,基于测量支付意愿的主流方法,结合客户调查法与联合分析法的关联水平系数思想,运用聚类的方法得出每个顾客的支付意愿；随后基于支付意愿、成本、库存、及消费者决策系数等因素建立了个性化定价模型,并且引入遗传算法,构造了使得企业目标利润最大化优化搜索算法,以此算法确定针对每一个客户的定价方案；最后分析了实施个性化定价的合理方案,对所确定的目标顾客实施个性化定价。本文的研究内容主要有：第一,分析了定价的概念及目标,总结了传统的定价方法,对个性化定价起源、类型及社会福利进行了分析,并指出在电子商务环境下进行个性化定价的挑战及机遇。第二,研究了个性化定价的一般步骤,根据该流程,研究了消费者支付意愿的测量方法及技术,并结合联合分析法的关联水平系数思想及客户调查思想提出适合本文研究内容的支付意愿测量方法。运用模糊c均值聚类算法对消费者进行聚类,以群体当中消费者购买历史中同类产品的最高价格作为该群体的支付意愿,以聚类结果隶属度作为每个个体隶属于每个群体的程度,以此程度作为消费者关联于各个群体支付意愿的水平系数,计算这个程度和每一个群体的支付意愿的乘积的累加和,便作为该客户的支付意愿。并收集消费者历史购物数据,采用本文算法测量其支付意愿,与直接询问值作对比分析。第三,基于消费者支付意愿,建立了关于产品成本、库存、消费者决策系数的个性化定价模型,并引入遗传算法。研究了遗传算法的原理,阐述了本文选择遗传算法的原因,构造了个性化定价的优化方法。根据实验所得支付意愿,运用遗传算法对所建模型进行优化搜索,通过大量实验确定遗传算法的参数,并对模型搜索结果进行了分析。第四,研究了实施个性化定价的具体方案,指出设置“门槛”为本文所建模型的最优选择,指出在对客户实施个性化定价时,先标出商品的价格,分析完通过“门槛”的客户信息,在客户到达或主动发放相应的折扣券给客户。
[Abstract]:Under the background of the date of demand of different customers, we have entered the era of personalized, through the analysis of consumers' personal information, to provide more products and services meet consumer preferences, especially in the e-commerce environment, the personalized service is more convenient, more favored by consumers. On the one hand, personalized products, electronic the business of personalized service, personalized recommendation and rapid development, but the current research on marketing of e-commerce personalized, tend to study personalized products and services recommended, and rarely involve individual price and promotion, make personalized pricing is not attractive. Over the same period of development, on the other hand, in recent years, with the the pursuit of individual people, the development of e-commerce personalized service, personalized pricing as an important part of personalized service, More and more attention has been paid to it. Coupled with the maturity of data mining technology, the industry is optimistic about the application prospect of personalized pricing in e-commerce environment.
This article from the consumer behavior, e-commerce product pricing environment and other aspects, analyzes the main steps of personalized pricing. For personalized pricing three steps: measuring willingness to pay, the search target customers, the implementation of personalized pricing, the mainstream method of measuring willingness to pay based on the combination of customer survey and analysis correlation coefficient method level the idea of using clustering method to obtain the willingness to pay for each customer; then based on the willingness to pay, cost, inventory, factors and consumer decision coefficient set up a personalized pricing model, and the introduction of genetic algorithm, constructed the enterprise profit maximization goal optimization search algorithm, this algorithm for determining the pricing scheme of every customer; the final analysis of the reasonable plan implementation of personalized pricing, the implementation of personalized pricing to determine the target customers. This research The main contents are as follows:
First, we analyze the concept and objectives of pricing, summarize the traditional pricing methods, analyze the origin, types and social welfare of personalized pricing, and point out the challenges and opportunities of personalized pricing under e-commerce environment.
Second, the general steps of personalized pricing, according to the flow, measuring method of consumer willingness to pay and the technology, and combined with the analysis of the correlation coefficient and the level of thought of customer survey proposed the willingness to pay of measurement for the content of this paper. The clustering of consumers using fuzzy C means clustering algorithm, with the highest price the group of consumers to buy similar products in the history as the willingness to pay for the group, with the clustering results of membership as individuals belonging to each group. This degree as consumers association coefficient in various groups of willingness to pay, and calculate the degree of willingness to pay for the product of each group of the accumulation and as the customer's willingness to pay and collect consumer shopping history data, measure the willingness to pay by this algorithm, and direct inquiry The value is compared and analyzed.
Third, based on consumer willingness to pay, based on the product cost, inventory, personalized pricing model of consumer decision-making coefficient, and the genetic algorithm is introduced. The principle of genetic algorithm, this paper expounds the reasons for choosing genetic algorithm, structural optimization method of personalized pricing. The willingness to pay according to the experimental result, using genetic algorithm to optimize the search the model parameters, the genetic algorithm is determined by experiments, and the model of search results were analyzed.
Fourth, study the specific implementation programs of personalized pricing, points out that establishing "threshold" to select the best model built in this paper, points out that in the implementation of personalized pricing to customers, marked the first commodity prices, after analyzing through the "threshold" of the customer information, arrival or active payment of the appropriate discount coupons to customers in the customer.

【学位授予单位】：华中师范大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F724.6;F274

【参考文献】