Cheng-Yang Lai

Identifying influential reviewers for word-of-mouth marketing

The key to word-of-mouth marketing is to discover the potential influential nodes for efficiently spreading product impressions. In this paper, a framework combined with mining techniques, a modified PMI measure, and an adaptive RFM model is proposed to evaluate the influential power of online reviewers. An artificial neural network is adopted to identify the target reviewers and a well-developed trust mechanism is utilized for effectiveness evaluation. This proposed framework is verified by the data collected from Epinions.com, one of the most popular online product review websites. The experimental results show that the proposed model could accurately identify which reviewers to select to become the influential nodes. This proposed approach can be exploited in effectively carrying out online word-of-mouth marketing, which can save a lot of resources in finding customers.

Discovering influencers for marketing in the blogosphere

Discovering influential bloggers will not only allow us to understand better the social activities taking place in the blogosphere, but will also provide unique opportunities for sales and advertising. In this paper, we develop an MIV (marketing influential value) model to evaluate the influential strength and identify the influential bloggers in the blogosphere. We analyze three dimensions of blog characteristics (network-based, content-based, and activeness-based factors) and utilize an artificial neural network (ANN) to discover potential bloggers. Based on peer and official evaluations, the experimental results show that the proposed framework outperforms two social-network-based methods (out-degree and betweenness centrality algorithms) and two content-based mechanisms (review rating and popular author approaches). The proposed framework can be effectively applied to support marketers or advertisers in promoting their products or services.

A social recommender mechanism for improving knowledge sharing in online forums

Nowadays, online forums have become a useful tool for knowledge management in Web-based technology. This study proposes a social recommender system which generates discussion thread and expert recommendations based on semantic similarity, profession and reliability, social intimacy and popularity, and social network-based Markov Chain (SNMC) models for knowledge sharing in online forum communities. The advantage of the proposed mechanism is its relatively comprehensive consideration of the aspects of knowledge sharing. Accordingly, results of our experiments show that with the support of the proposed recommendation mechanism, requesters in forums can easily find similar discussion threads to avoid spamming the same discussion. In addition, if the requesters cannot find qualified discussion threads, this mechanism provides a relatively efficient and active way to find the appropriate experts.

Identifying bloggers with marketing influence in the blogosphere

Finding influential bloggers will not only allow us to better understand interesting activities happening in a social network, but also present unique opportunities for sales and advertisements. In this paper, we address a novel problem of finding influential bloggers with marketing value in the blogosphere by proposing a MIV (Marketing Influential Value) model. We induce two dimensions of blog characteristics (network-based factors and content-based factors) and develop an adaptive artificial neural network (ANN) to identify the potential bloggers with marketing influence to support the marketers or advertisers in promoting their products or services effectively.

Building a qualitative recruitment system via SVM with MCDM approach

Advances in information technology have led to behavioral changes in people and submission of curriculum vitae (CV) via the Internet has become an often-seen phenomenon. Without any technological support for the filtering process, recruitment can be difficult. In this research, a method combining five-factor personality inventory, support vector machine (SVM), and multi-criteria decision-making (MCDM) method was proposed to improve the quality of recruiting appropriate candidates. The online questionnaire personality testing developed by the International Personality Item Pool (IPIP) was utilized to identify the personal traits of candidates and both SVM and MCDM were employed to predict and support the decision of personnel choice. SVM was utilized to predict the fitness of candidates, while MCDM was employed to estimate the performance for a job placement. The results show the proposed system provides a qualified matching according to the results collected from enterprise managers.

A social referral appraising mechanism for the e-marketplace

Abstract With the popularity of social media, in order to achieve word-of-mouth marketing, many marketers and individual sellers are actively engaged in achieving positive ratings in the e-marketplace. Although many seller review mechanisms have been developed, e-commerce market operators and consumers still face trust fraud issues. Trust is one of the major issues that confuse online purchase activities. In this paper, we develop a social referral mechanism to verify sellers based on the experiences of friends within a buyer’s social network. The proposed framework estimates the trust values of sellers from the perspective of the social network in order to stop buyers from making transactions with fraudulent sellers in the online marketplace.

A diffusion planning mechanism for social marketing

Abstract Social media is gaining importance as a component of marketing strategies. Many types of social media, such as social networking sites, blogospheres and micro-blogospheres, have been seeking business opportunities and establishing brand expression in the recent years. Online marketing information diffusion has become the critical business model of online social networks. However, most of the current marketing studies on discovering potential influences do not appropriately support them to diffuse advertisements. Therefore, marketing information may be lost during the diffusion process and cannot be sent to potential customers successfully. In this research, a diffusion path planning mechanism for advertisements is developed to help influencers to propagate marketing information and help marketers to evaluate possible rewards under different marketing strategies. Our experimental results show that the proposed mechanism can significantly improve the diffusion process of advertising messages and decrease marketing uncertainty.

A Diffusing Path Planning Mechanism for Marketing Information Propagation over Social Media

Social media are gaining importance as a component of marketing strategies. Many of them, such as social networking site, blogosphere, and micro-blogosphere, have been seeking business opportunities and establishing brand expression in recent years. Online marketing information diffusion becomes the critical business model for online social networks. Marketers attempt to diffuse advertisements for reaching potential customers through the Internet. However, current marketing research discovered possible influencers but not appropriate support them to diffuse advertisements. In this research, a diffusing path planning mechanism for advertisement was developed for supporting influencers to propagate marketing information and supporting marketers to evaluate possible reward under different marketing strategies.

Business modeling for online video services: download vs. streaming

Download and streaming are two major approaches for delivering online video/movie. In this paper, considering their differentiated characteristics, we analyze the pricing issues in a market where both types of content distribution are available. We examine the impact of various technological and market factors on developing the business strategies under the scenarios that these distribution channels are owned by the same provider or two independent firms. We further examine the loyalty fee development when the video content is independently owned by another content provider. Our results shows that while the quality of two types of video distribution has significant impact on determining the content loyalty fee in a competitive market, loyalty fee is only affected by the characteristic of steaming service.

A Social Referral Mechanism on e-Marketplace

With the popularity of social media, in order to achieve the word-of-mouth marketing, many marketers and individual sellers are actively engaged in getting positive ratings on e-marketplace. Although there are many seller review mechanisms developed, the electronic commerce market operators and the consumers are still facing the trust fraud challenge. Trust is one of major issues that baffle online purchase activities. In this paper, we develop a seller referral mechanism to verify sellers based on the experience of friends within a buyer’s social network. The proposed framework estimates trust values of sellers from the perspective of social network for helping buyers prevent making transactions with the fraudulent sellers in the online marketplace.