Search

Predicting customer repurchase propensity/frequency has received broad research interests from marketing, operations research, statistics, and computer science. In the field of marketing, Buy till You Die (BTYD) models are perhaps the most representative techniques for customer repurchase prediction. Those probabilistic models are parsimonious and typically involve only recency and frequency of customer activities. Contrary to BTYD models, a distinctly different class of predictive models for customer repurchase is machine learning. This class of models include a wide variety of computational and statistical learning algorithms. Unlike BTYD models built on low-dimensional inputs and behavioral assumptions, machine learning is more data-driven and excels at fitting predictive models to a large array of features from customer transactions. Using a large online retailing data, we empirically assess the prediction performance of BTYD modeling and machine learning. More importantly, we investigate how the two approaches can complement each other for repurchase prediction. We use the BG/BB model given the discrete and non-contractual problem setting and incorporate BG/BB estimates into high-dimensional Lasso regression. In addition to showing significant improvement over BG/BB and Lasso without BG/BB, the integrated Lasso-BG/BB provides interpretability and identifies BG/BB predictions as the most influential feature among ∼100 predictors. The lately developed CART-artificial neural networks exhibit similar patterns. Robustness checks further show the proposed Lasso-BG/BB outperforms two sophisticated recurrent neural networks, validating the complementarity of machine learning and BTYD modeling. We conclude by articulating how our interdisciplinary integration of the two modeling paradigms contributes to the theory and practice of predictive analytics.

Blockchain technology has the promise of transforming security and trust in digital transactions. However, concerns about technical complexity and the benefits of deployment have blunted its adopti...

PurposeThis paper reports the panel discussion on the topic of artificial intelligence (AI) and robots in our lives. This discussion was held at the Digitization of the Individual (DOTI) workshop at the International Conference on Information Systems in 2019. Three scholars (in alphabetical order: Ting-Peng Liang, Lionel Robert and Suprateek Sarker) who have done AI- and robot-related research (to varying degrees) were invited to participate in the panel discussion. The panel was moderated by Manuel Trenz.Design/methodology/approachThis paper introduces the topic, chronicles the responses of the three panelists to the questions the workshop chairs posed and summarizes their responses, such that readers can have an overview of research on AI and robots in individuals' lives and insights about future research directions.FindingsThe panelists discussed four questions with regard to their research experiences on AI- and robot-related topics. They expressed their viewpoints on the underlying nature, potential and effects of AI in work and personal life domains. They also commented on the ethical dilemmas for research and practice and provided their outlook for future research in these emerging fields.Originality/valueThis paper aggregates the panelists' viewpoints, as expressed at the DOTI workshop. Crucial ethical and theoretical issues related to AI and robots in both work and personal life domains are addressed. Promising research directions to these cutting-edge research fields are also proposed.

With the support of smart technology, IT-enabled services have become “smart” and have progressively disrupted existing markets. Ride-hailing services (RHSs) are widely regarded as representative of these IT-enabled services. However, few studies on IT-enabled services investigate how the technological attributes of smart technology influence service performance in a continuously changing environment. We developed our research model according to Wixom and Todd’s model, the literature on change management, and the literature on information system postadoption behavior. We conducted a large-scale field study by surveying 380 drivers from major metropolises in mainland China and a post hoc qualitative interpretation to validate our model. We found that smart technological attributes of RHS systems (i.e., monitoring, control, advisory support, and responsive support) positively influence functionality and content quality, which in turn influence service quality. In addition, service quality positively influences drivers’ postadoption attitudes and behaviors, including openness to RHS change, job satisfaction, and continuous usage intentions. Our findings provide important theoretical and practical implications.