电子商务环境下基于大数据的情景定价与产品组合推广机制研究

An e-commerce platform is believed to increase awareness of niche products by long-tail effects. However, the fast expansion of the e-commerce platform leads to information overload; as a result, many niche products are quickly lost in the colossal product group and are much harder to be noticed by consumers. Niche products are necessary for an e-commerce platform to maintain diversity, but the current information overload is not good for promoting new niche products, and will even undermine the further development of the e-commerce platform. .To establish a sound balance among the size of the e-commerce platform, product diversity, and consumers’ faith, this research proposes a big-data based transactional mode and pricing mechanism for a promotion mix consisting of popular products and niche products on an e-commerce platform. The main contribution of this research lies in the following three aspects: .1. This research proposes a new method to match products and characteristics of consumers’ buying behaviors. .2. Based on the aforementioned method, this research designs a novel transactional mode for joined promotion of popular products and niche products..3. Furthermore, this research proposes a contextual pricing mechanism based on the combination of products and multiple characteristics of consumers’ buying behaviors. .The results of this research will be beneficial for improving the transactional efficiency and credibility between consumers and products, maintaining product diversity on an e-commerce platform, and utilizing abundant information resources to optimize product pricing strategies.

传统观点认为电子商务平台能够通过“长尾效应”拓展利基产品的受众面，然而，电商平台规模扩张带来的信息过载导致众多利基产品淹没在日益膨胀的产品群中，被消费者发现的可能性不增反降。利基产品是保障平台多样性的必要条件，然而当前利基产品易逝的现状不利于利基产品的推陈出新，为电子商务平台的进一步发展埋下隐患。为了在市场规模、平台多样性和消费者信心之间建立起良好的平衡，本研究提出一种基于大数据的电子商务市场“热门”产品和“利基”产品组合推广的交易模式及定价机制。研究的主要贡献包括：1、提出向产品匹配电商平台终端消费者购买行为特征的方法；2、以此为基础，建立新的“热门”产品与“利基”产品联合推广和交易模式；3、在以上研究基础上，构建基于产品和多维度消费者购买行为特征的情景定价机制。研究成果有助于提高用户与产品之间建立交易的效率和信任性，维持电商平台的多样性，有利于充分利用丰富的信息资源优化产品定价策略。

项目自2016年执行以来，课题组紧紧围绕“电子商务环境下基于大数据的情景定价与产品组合推广研究”开展研究。主要分为“情景数据信息抽取与挖掘”和“用户行为模式识别与机理分析”两个研究方面。第一方面的研究成果提供了一套基于文本分析与机器学习技术的海量情景数据抽取与挖掘方法，能够高效地抽取用户决策行为的情景信息，为有效分析用户行为模式及产生机理提供信息技术保证。另一方面的研究成果主要体现在准确识别用户决策行为进而能有效地分析用户行为产生的机理，从而为企业实施优化策略提供可行空间。截至2019年底，共发表（含被接收）标注基金号资助的论文30篇，期刊论文16篇（其中Dallas24期刊论文2篇、SCI/SSCI论文8篇），会议论文14篇（其中EI检索11篇）；授权发明专利1项，研究成果较为丰富。