Research on Online Word-of-Mouth Spread Mechanism based on the Node Properties and Word-of-Mouth Type
Min Zhao,Xiaodong Qian,Rui-Dian Zhang
DOI: https://doi.org/10.2991/ICMESD-18.2018.137
2018-05-01
Abstract:In this paper, we studied that how node properties and word-of-mouth type influenced the spread of word of mouth. First of all, according to previous research, we embedded all kinds of influencing factors like word-of-mouth type and customer recognition in the word-of-mouth spread model on the basis of SIR virus propagation model. Then, we made a simulation with the word-of-mouth model. According to the simulation, we could draw a graph of node density to examine how these factors affected the change of the node density, which reached a spread mechanism based on the principle and path of word-of-mouth spread. Finally, we give some practical proposals for enterprise marketing. Introduction Use the complex network as the framework to study mechanism of word-of-mouth communication. From the point of view of word-of-mouth type, scholars such as Ye [1] have conducted research on online reviews and merchant credibility of China's C2C e-commerce site --Taobao. The results show good reviews and good business reputation has a positive effect on the increase in sales revenue of merchants. Park [2] (2008) studied the influence of the quality, direction and type of Internet word-of-mouth on the effectiveness of Internet word-of-mouth communication. From the perspective of customer awareness, Li Dongjun [3] proposed the interaction between word of mouth and customer perception from the perspective of customer awareness. From the perspective of network structure parameters, Styer et al. [4] (2006) studied online forum word-of-mouth communication networks. The distribution of degree degrees obeys the power law distribution rather than the Poisson distribution previously considered. Yang Yanyang [5] proposed that when the node with high absolute value of recognition has the characteristics of high topology, it has the greatest impact on the network. Li Lin, Sun Jun, and Zhou Zhangjin[6] have studied the heterogeneity of nodes and the difference in network topology to bring about differences in propagation probabilities and propagation ranges. The SIR model was used as the framework of word-of-mouth communication, embedded in the large-scale online social network-micro blogging data, using MATLAB, Pajek to simulate the word-of-mouth communication. The Establishment of a Model Information Dissemination Model The user density is including: s(t) in susceptible state, i(t) in infected state, r(t) in immune state. Due to the highly heterogeneous nature of online networks, online social networks are often abstracted as complex networks.[7]. In general, network information dissemination can be characterized by the SIR model, that is, an infected unknown person can become a communicator with a certain probability and become an immune communicator after a certain period of time [8]. Zhang Haifeng [9] used the virus propagation framework and the connection probabilities between nodes to construct the information propagation model in his research. 792 Copyright © 2018, the Authors. Published by Atlantis Press. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/). 4th Annual International Conference on Management, Economics and Social Development (ICMESD 2018) Advances in Economics, Business and Management Research (AEBMR), volume 60
Computer Science,Business