社会化商务中基于多重关系的社会网络形成机制及其对产品销售的影响

发布时间：2018-06-25 15:44

本文选题：社会化商务 + 社会网络　；参考：《武汉大学》2015年博士论文

【摘要】：本文主要从三个部分来系统地探讨了在社会化商务背景下,基于用户之间多重关系的社会网络虚拟资源的形成及其商业价值和相应的营销策略,在具体的研究中,我们选取社会化商务性质较强的交易型社区为研究对象,探讨用户在其中形成的多重关系以及这些社会化关系的商业价值并最终帮助企业探讨有效的营销策略来利用这些虚拟社会资源。本文的第一部分从社会网络演化的基本动力出发,探讨在不同动机的驱动下,社会化商务中的多重关系网络的形成和演化,研究结果表明,由于交易型社区中同时存在着买家和卖家,社区中成员之间关系的构建会受到不同机制的影响。从买家的角度来看,当交易型社区中的买家与社区中的其他买家或卖家构建关系时,他们主要受到观察学习和社会传染两种机制的影响；从卖家的角度来看,同质性,互惠性以及结构等价则是影响他们发出连接关注社区中的买家和卖家的主要因素。本文的第二部分通过向量自回归模型探讨了交易型社区中用户相互之间社会网络关系的形成对社区中的卖家商品交易行为的动态关系。研究结果表明,交易型社区中的卖家在收到来自买家或卖家的关注时,其销售收入都会得到显著的提升,而且这些积极的影响会在关系形成后的第四天达到顶峰；而相比之下,不论是买家还是卖家在关注交易型社区中的其他买家时,并不会显著地影响到社区内的销售收入。另外,通过脉冲响应函数,我们还可以看出不同类型关系的形成对销售收入的长期和短期影响,我们发现在交易型社区中,卖家与卖家之间相互构建关系时,从长期来看会给社区的整体销售收入带来最强的影响(1.306),其长期累积的显著影响比买家与卖家构建的关系的长期影响(0.327)高出了近4倍；相比之下,买家与其他买家以及卖家与买家构建的关系则并不具有显著的影响。最后,在第三部分的研究中,我们通过基于元胞自动机的仿真方法探讨了企业如何利用交易型社区中的虚拟社会资源来设计具体的营销策略来提升市场绩效。基于ABMS的仿真实验结果表明,信息在基于社会影响机制的扩散与基于同质性影响的扩散存在着显著的差异性,并且在控制了网络关系密度等因素的影响下,这种差异性仍然存在；而当我们通过对真实网路关系中的关系结构进行随机抽样而使其符合真实网络的结构特性时,社会影响机制与同质性影响机制的差异性依然显著。具体来讲,同质性影响机制能够在病毒式营销策略推出的早期阶段迅速提升扩散传播的影响范围,而随着扩散时间的推移,由于整个社区的发帖或回复数量有限,这也导致同质性机制的影响并不能一直保持快速的增长。另一方面,社会影响机制则不受具体社区活动的限制,因此随着扩散时间的增长,该机制能在更大范围内将特定的产品或服务相关信息扩散到社区中的每一个用户中去。该研究也针对传播扩散初始节点类型的选择做出了相应的分析,研究结果表明,传统的按照单独的人际网络所确定的中心人物作为初始节点传播的策略在交易型社区中并不是最优的策略选择,通过选择处于人际网络和事件归属网络双重网络的综合中心位置的用户实施病毒式营销策略时,结果相对于随机选择的用户对最终的传播扩散效果具有显著的提升作用。最后,通过基于真实网络传播数据的生存分析,我们进一步证明了同质性传播机制在交易型社区中相对于社会影响传播机制具有更加显著的作用。
[Abstract]:This paper systematically discusses the formation, commercial value and marketing strategy of social network virtual resources based on multiple relationships among users in the context of socialized business. In the specific study, we select the socialized business community with strong commercial nature as the research object, and discuss the user in it. The multiple relations and the commercial value of these socialized relationships finally help the enterprises to explore effective marketing strategies to use these virtual social resources. The first part of this paper, starting from the basic motive force of social network evolution, explores the formation and the formation of multiple relational networks in socialized business under the drive of different motives. Evolution, the results show that the construction of relationships among members in the community is affected by different mechanisms due to the presence of buyers and sellers in the trading community. From the buyer's point of view, when buyers in the trading community build relationships with other buyers or sellers in the community, they are mainly subject to observation learning and social transmission. From the point of view of the seller, from the point of view of the seller, homogeneity, reciprocity and structural equivalence are the main factors affecting the buyers and sellers in the community of connection and concern. The second part of this paper explores the formation of social network relationships among the users in the trading community by the vector autoregressive model to the community. The results show that the sales income of the sellers in the trading community is significantly improved when they receive the attention of the buyers or sellers, and the positive effects will peak in the fourth day after the relationship is formed; in contrast, both the buyer and the seller are in the market. Paying attention to other buyers in a trading community does not significantly affect the sales income in the community. In addition, through the impulse response function, we can see the long-term and short-term effects of the formation of different types of relationships on sales revenue. We find that in the transaction type community, the relationship between the seller and the seller is long. The period will have the strongest impact on the overall sales revenue of the community (1.306), and its long-term cumulative impact is nearly 4 times higher than the long-term impact of the relationship between buyers and sellers (0.327); by contrast, the relationship between buyers and other buyers, as well as sellers and buyers, is not significantly affected. Finally, in the third part In the study, we discuss how enterprises use virtual social resources in the trading community to design specific marketing strategies to improve market performance by using the simulation method based on cellular automata. The results of simulation experiments based on ABMS show that the diffusion of information based on social impact mechanism and the influence of homogeneity based on the social impact mechanism exist. There are significant differences, and the difference still exists under the influence of controlling the network relationship density, and the difference between the social impact mechanism and the homogeneity influence mechanism is still significant when we random sampling the relationship structure in the real network relationship to make it conform to the real network structure characteristics. Specifically, the homogeneity impact mechanism can rapidly increase the impact of diffusion on the early stage of the viral marketing strategy. With the passage of time, the number of posts or replies of the whole community is limited, which leads to the fact that the impact of homogeneity mechanism does not maintain rapid growth. The impact mechanism is not restricted by specific community activities, so with the growth of diffusion time, the mechanism can spread specific product or service related information to every user in the community in a wider range. It is not the best strategy in the transaction type community that the central character identified by the individual human network as the initial node is not the best choice in the transaction type community. The result is compared to the random selection by selecting the users who are in the integrated center position of the human network and the event home network. Finally, through the survival analysis based on the real network data, we further prove that the homogeneity transmission mechanism has a more significant role in the transaction community relative to the social impact transmission mechanism.
【学位授予单位】：武汉大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：F274

【引证文献】