Network promoter score (NePS): An indicator of product sales in E-commerce retailing sector

2022 | Electronic Markets | Citations: 0

Authors: Mandal, Supriyo; Maiti, Abyayananda

Abstract: E-commerce companies want to predict their future product sales from the current ... Expand

Abstract: E-commerce companies want to predict their future product sales from the current customers’ feedback to frame a better business strategy. However, the conventional way of analyzing rating activities or quality and sentiment of reviews, volume of sales, or product prices is not enough for establishing a strong regression between these parameters and future product sales. Most of the existing works ignore the heterogeneous positional and influential effects of individual customer reviews and ratings. For the realization of these effects, we use review network i.e., a bipartite network between customers and products based on the customers’ review activities. In this paper, we present a concept named Network Promoter Score (NePS) based on the reliability, positional influence of each customer in the network. In-depth experiments on online review datasets show that NePS emerges as a strong indicator of product sales and can be remarkably futuristic compared to the existing parameters. Furthermore, we propose a predictive modeling technique to estimate the product sales of a company based on NePS. Collapse

Topics: mobile phone user-generated content online review electronic commerce video game

Methods: experiment cluster analysis sentiment analysis machine learning deep learning

Developing a Composite Measure to Represent Information Flows in Networks: Evidence from a Stock Market

2022 | Information Systems Research | Citations: 2

Authors: Shangguan, Wuyue (Phoebe); Leung, Alvin Chung Man; Agarwal, Ashish; Konana, Prabhudev; Chen, Xi

Abstract: There is increasing interest in information systems research to model informatio ... Expand

Abstract: There is increasing interest in information systems research to model information flows from different sources (e.g., social media, news) associated with a network of assets (e.g., stocks, products) and to study the economic impact of such information flows. This paper employs a design science approach and proposes a new composite metric, eigen attention centrality (EAC), as a proxy for information flows associated with a node that considers both attention to a node and coattention with other nodes in a network. We apply the EAC metric in the context of financial market where nodes are individual stocks and edges are based on coattention relationships among stocks. Composite information from different channels is used to measure attention and coattention. To evaluate the effectiveness of the EAC metric on predicting outcomes, we conduct an in-depth performance evaluation of the EAC metric by (1) using multiple linear and nonlinear prediction methods and (2) comparing EAC with a benchmark model without EAC and models with a set of alternative network metrics. Our analysis shows that EAC significantly outperforms other measures in predicting the direction and magnitude of abnormal returns of stocks. Besides, our EAC specification has better predictive performance than alternative specifications, and EAC outperforms direct attention in predicting abnormal returns. Using the EAC metric, we derive a stock portfolio and develop a trading strategy that provides significant and positive excess returns. Lastly, we find that composite information has significantly better predictive performance than separate information sources, and such superior performance owes to information from social media instead of traditional media. Collapse

Topics: social media evaluation criteria electronic trading database system accounting

Methods: machine learning design science longitudinal research autoregressive conditional heteroscedasticity model random forest classification

Deriving Execution Effectiveness of Crowdfunding Projects from the Fundraiser Network

2022 | Journal of Management Information Systems | Citations: 3

Authors: Kao, Ta-Wei; Hsiao, Shih-Hui; Su, Hung-Chung; Ku, Chih-Hao

Abstract: Reward-based crowdfunding has become a popular fundraising marketspace for entre ... Expand

Abstract: Reward-based crowdfunding has become a popular fundraising marketspace for entrepreneurs. However, few studies take a process perspective and examine how fundraiser network influences the execution effectiveness of crowdfunding projects. Collecting the funding activities of all fundraisers on Kickstarter in 2017, we construct a fundraiser network and examine the effects of fundraiser network structural characteristics on project execution effectiveness. Employing endogenous stochastic frontier analysis, we show that reciprocity and accessibility in the fundraiser network increase project execution effectiveness. We also demonstrate that gaining network status by virtue of being an authoritative figure in the fundraiser network increases project execution effectiveness. In sum, our study identifies the effects of the fundraiser network characteristics on project execution effectiveness and introduces a process perspective for evaluating crowdfunding campaigns. Collapse

Topics: crowdsourcing crowdfunding social capital Facebook social media

Methods: instrumental variables estimation chi squared test post-hoc analysis computational algorithm theoretical contribution

Theories: institutional theory

News-Induced Dynamic Networks for Market Signaling: Understanding the Impact of News on Firm Equity Value

2021 | Information Systems Research | Citations: 0

Authors: Chen, Kun; Li, Xin; Luo, Peng; Zhao, J. Leon

Abstract: Firm relations, which inform the competitive environments of firms, are critical ... Expand

Abstract: Firm relations, which inform the competitive environments of firms, are critical to firm operations and are often factored into investor decisions. Previous studies and practices have considered relatively stable and long-term business relations as an indicator of firm market value. Public news often reports on business relations, especially dynamic and short-term opportunities and challenges involving different partners. Learning about firm relations from news is commonly done by human investors but has not been studied systematically in previous research, leading to a research opportunity in market signaling via dynamic firm relations. In this study, we propose a new approach to market signaling by leveraging text-mining methods to extract cobenefit/counter-benefit networks based on firms’ mutual or conflicting interests in business events. We find that the resulting networks in the long term are partially aligned with firm cooperation and competition relationships, confirming their semantic implications for investor perception and attention. Our empirical study further shows that dynamic networks formed in a short time period (measured by firm centrality in networks) have significant impacts on firm equity value, after controlling for market activities and other information from news, such as volume, sentiment, and comentions. We show that dynamic networks can provide additional value in predicting firm equity value over stable networks. Moreover, the negative effects of counter-benefit networks emerge rapidly and persist longer than the positive effects of cobenefit networks. This study provides new insights into investor perception of news and suggests new research directions for financial text mining. Our research findings on market signaling via news-induced networks also have an impact on financial practices, such as market analysis and automatic trading. Collapse

Topics: network effect database system supply chain management blog accounting

Methods: computational algorithm machine learning experiment longitudinal research natural language processing

Theories: cognitive attention theory

On The Role Of Centrality In Information Diffusion In Social Networks

2013 | European Conference On Information Systems | Citations: 1

Authors: Mochalova, Anastasia; Nanopoulos, Alexandros

Abstract: Towards understanding how people use social media to interact with each other, i ... Expand

Abstract: Towards understanding how people use social media to interact with each other, it becomes important to investigate the influence mechanism that emerges in online social networks. Due to it, information in social networks often diffuses “virally”. In this paper, we recognize the relation between the influence of the seed members that trigger information diffusion and the attitude of the rest members towards the information. To characterize the influence of seed members, we examine their centrality within the network structure, based on the premise that more central members can reach many other members. We examine a comprehensive set of centrality scores for identifying the influence of seed members and conduct a thorough experimental evaluation with data from a real social network. Our experimental results provide insights into the interplay of the various centrality scores with the members’ attitude and how this interplay affects the outcome of information diffusion. The first contribution of the proposed approach is, thus, a solution to the influence maximization problem that is based only on information about the structure of a social network and does not require any additional knowledge about the quantification of influence between its members. The second contribution is a thorough investigation of the performance of the various centrality-based seedselection methods with respect to factors such as the seed size and members’ attitude, indicating that different centrality scores are suitable in different cases. Collapse

Topics: social network consumer behavior social media Twitter Facebook

Methods: descriptive statistic experiment literature study computational algorithm survey