Research Statement

Last update: Jan 25, 2022.

The objective of my research is to advance Business Analytics using Artificial Intelligence (AI) thereby creating new theoretical knowledge and business value from big data. With the advent of IT platforms (e.g., corporate reports, social media, mobile platforms, e-commerce, IoT, robots), we are witnessing an unprecedented amount of data capturing consumer behaviors, business operations, and societal dynamics. Recent algorithmic advances in AI (e.g., natural language processing, machine learning, computer vision, deep learning) and the availability of powerful computational resources (e.g., GPU, cloud computing) allow us to effectively and efficiently analyze big data, uncover business insights, and make data-driven decisions. This new phenomenon of business analytics creates opportunities and challenges for academic researchers and industry practitioners. The main contributions of my research are in (i) the methodological development of data analytics and (ii) the applications of data analytics and AI in data-intensive IT platforms with the following themes:

1. Development of AI-based analytics methods [MISQ 2016MISQ 2020; CIST 2019; WITS 2020; WITS 2021; IMC 2005; INFOCOM 2013; PAM 2015]: According to industry reports, 80-90% of the data are in unstructured formats (e.g., text, image, audio, video). These datasets provide a great opportunity to analyze detailed and nuanced information on consumer behavior, business operations and strategies, and societal phenomena. However, they also pose a methodological challenge in incorporating such data into conventional research models. This line of research is to develop analytics methods and frameworks using a variety of AI techniques (e.g., natural language processing (NLP), computer vision, machine learning, deep learning) to quantify business decisions and their outcomes. These approaches allow researchers to construct a variety of variables that are otherwise impossible to operationalize.

2. Studying IT platforms with analytics and AI approaches [JMIS 2016ISR 2020ISR 2021; WITS 2020; HICSS 2020; WITS 2020; HICSS 2021; WITS 2021; WISE 2021; SECON 2013]: While the first research stream focuses on methodological innovations, this line of work emphasizes the applications of analytics and AI methods to derive new theoretical insights from data-intensive IT platforms (e.g., social networks, mobile app markets, e-commerce, video platforms). Specifically, with detailed activities of consumers and firms captured in the IT platforms, I use analytics and AI to study a variety of interactions among them (e.g., friendship formation, cross-promotion matching, cross-product review effect, video ad placement).

3. Mitigating IT risk with analytics [JC 2016; JMIS 2020; CIST 2017; HICSS 2020]: Our dependence on IT and analytics can come with unintended consequences for cybersecurity.  This line of work is to use big data analytics approaches to understand and mitigate cyber risks. Specifically, we collect and analyze large-scale cyber risk datasets, including spam emails, phishing websites, risk disclosure documents, and data breach events. Then we conduct randomized experiments and econometric analyses to measure the causal impact of security information disclosure on firms’ security strategies and investors’ responses.

4. Interdisciplinary big data analytics research [JBE 2019; CFMA 2019; APIOC 2019]: The application of analytics and AI goes beyond MIS research. I often collaborate with researchers from a variety of research domains, including accounting, marketing, organization science, technology management, economics, and mathematics.

In summary, my research bridges AI and business analytics research. With the proliferation of IT platforms, the amount and complexity of business data will only increase. Thus, business analytics research should be able to propose innovative and practical methods to extract business value from datasets whether they are in structured or unstructured formats. As shown in my research publications, AI approaches (e.g., NLP, computer vision, and deep learning) have shown to create significant business value by solving “perception” problems (e.g., understanding text documents and recognizing semantics from images). By leveraging other AI approaches, such as reinforcement learning, planning, and knowledge representation, I plan to expand my AI-centric business analytics research to enable high-level “reasoning” tasks in business decision-making.