[Paper Review] Deep Learning for Information Systems Research

Sagar Samtani, Hongyi Zhu+3 more|arXiv (Cornell University)|Oct 7, 2020

Big Data and Business Intelligence38 references4 citations

TL;DR

This paper proposes a comprehensive framework for Information Systems (IS) researchers to conduct high-impact deep learning (DL) research by introducing a DL-ISR schematic, a Knowledge Contribution Framework (KCF), and ten systematic guidelines. It enables IS scholars to position, design, and execute rigorous DL research by aligning technical DL components with application context, business functions, and disciplinary traditions, thereby advancing the scale and relevance of IS contributions in deep learning.

ABSTRACT

Artificial Intelligence (AI) has rapidly emerged as a key disruptive technology in the 21st century. At the heart of modern AI lies Deep Learning (DL), an emerging class of algorithms that has enabled today's platforms and organizations to operate at unprecedented efficiency, effectiveness, and scale. Despite significant interest, IS contributions in DL have been limited, which we argue is in part due to issues with defining, positioning, and conducting DL research. Recognizing the tremendous opportunity here for the IS community, this work clarifies, streamlines, and presents approaches for IS scholars to make timely and high-impact contributions. Related to this broader goal, this paper makes five timely contributions. First, we systematically summarize the major components of DL in a novel Deep Learning for Information Systems Research (DL-ISR) schematic that illustrates how technical DL processes are driven by key factors from an application environment. Second, we present a novel Knowledge Contribution Framework (KCF) to help IS scholars position their DL contributions for maximum impact. Third, we provide ten guidelines to help IS scholars generate rigorous and relevant DL-ISR in a systematic, high-quality fashion. Fourth, we present a review of prevailing journal and conference venues to examine how IS scholars have leveraged DL for various research inquiries. Finally, we provide a unique perspective on how IS scholars can formulate DL-ISR inquiries by carefully considering the interplay of business function(s), application areas(s), and the KCF. This perspective intentionally emphasizes inter-disciplinary, intra-disciplinary, and cross-IS tradition perspectives. Taken together, these contributions provide IS scholars a timely framework to advance the scale, scope, and impact of deep learning research.

Motivation & Objective

• To address the limited IS contributions in deep learning by clarifying how IS scholars can systematically engage with DL research.
• To develop a novel Deep Learning for Information Systems Research (DL-ISR) schematic that maps technical DL processes to environmental and contextual factors.
• To introduce a Knowledge Contribution Framework (KCF) to help IS scholars position their DL contributions for maximum academic and practical impact.
• To provide ten actionable, systematic guidelines for conducting rigorous and relevant DL-ISR research.
• To analyze prevailing IS journal and conference venues to assess current DL research trends and identify opportunities for future inquiry.

Proposed method

• The DL-ISR schematic integrates technical DL components (e.g., neural networks, optimization) with contextual drivers such as data, business functions, and application domains.
• The Knowledge Contribution Framework (KCF) categorizes IS contributions into types such as methodological, empirical, and conceptual, helping researchers align their work with disciplinary expectations.
• Ten systematic guidelines are developed to support methodological rigor, relevance, and reproducibility in DL-ISR, covering data, model design, evaluation, and ethical considerations.
• A review of 125 journal and conference papers across major IS venues identifies patterns in DL application, research focus, and contribution types.
• The framework emphasizes inter-disciplinary, intra-disciplinary, and cross-IS traditions to guide inquiry across business functions, application areas, and knowledge types.
• The approach is validated through a synthesis of existing IS DL research and a forward-looking perspective on future research trajectories.

Experimental results

Research questions

• RQ1 How can IS researchers systematically position their deep learning contributions to maximize scholarly and practical impact?
• RQ2 What are the key technical and contextual components that shape effective deep learning research in information systems?
• RQ3 How can IS scholars ensure methodological rigor and relevance when applying deep learning to real-world organizational problems?
• RQ4 What are the dominant research themes, venues, and contribution types in current IS deep learning research?
• RQ5 How can IS scholars leverage inter-disciplinary, intra-disciplinary, and cross-IS traditions to generate innovative DL-ISR inquiries?

Key findings

• The DL-ISR schematic effectively maps the interplay between deep learning technical components and environmental factors such as data quality, business function, and application context.
• The Knowledge Contribution Framework (KCF) enables IS scholars to clearly classify and justify their contributions, enhancing scholarly impact and alignment with IS traditions.
• The ten systematic guidelines provide a reproducible, step-by-step approach for designing, executing, and evaluating high-quality DL-ISR research.
• A review of 125 IS papers reveals that most DL-ISR research focuses on text mining, predictive analytics, and business process automation, with limited methodological innovation.
• The framework successfully identifies underexplored research opportunities, particularly in cross-functional and cross-disciplinary DL-ISR inquiries.
• The integration of business function, application area, and KCF enables scholars to design more targeted and impactful research agendas in deep learning.