Decision Support Systems最新文献

Live streaming commerce thrives with a rich tapestry of multimodal information that intertwines with various entities, including the anchor, the commodities, and the live streaming environment. Despite the wealth of data at hand, the synthesis and analysis of this information to predict sales remains a significant challenge. This study introduces a framework for multi-entity multimodal fusion, which is characterized by the effective synthesis of multimodal data and its prioritization of entity-level fusion, thereby providing a comprehensive feature representation for improving predictive performance. In addressing the multimodal data associated with a diverse range of products, our framework improves the Transformer architecture to initially capture the intra-product modal features and subsequently integrate the inter-product features. Data experiments are conducted on a real-world dataset from Taobao Live. The framework outperforms both traditional machine learning methods and state-of-the-art multimodal fusion methods, which affirms its value as a robust decision-support tool for sales prediction, enabling more accurate pre-event predictions and strategic planning. We also examine the impact of different types of information in accurate sales prediction. It is found that harnessing a comprehensive suite of data leads to optimal performance across all evaluation metrics. Commodity-related data is primary factor in determining the prediction accuracy, followed by video data and streaming room-related data, providing insights regarding the resource allocation for collecting and analyzing multimodal data from live streaming platforms.

This paper contextualizes explainable artificial intelligence (AI) for enhanced decision-making and serves as an editorial for the corresponding special issue. AI is defined as the development of computer systems that are able to perform tasks that normally require human intelligence by understanding, processing, and analyzing large amounts of data. AI has been a dominant domain for several decades in the information systems (IS) literature. To this end, we define explainable AI (XAI) as the process that allows one to understand how an AI system decides, predicts, and performs its operations. First, we contextualize its current role for improved business decision-making. Second, we discuss three underlying dimensions of XAI that serve as broader innovation grounds to make better and more informed decisions, i.e., data, method, and application. For each of the contributing papers in this special issue, we describe their major contributions to the field of XAI for decision making. In conclusion, this paper further presents a future research agenda for IS researchers in the XAI field.

The emergence of online healthcare platforms has changed the competitive environment among physicians. However, little is known about how physicians can improve their performance in this new environment. Platforms also face challenges in comprehending the competitive mechanisms among physicians, which might hinder them from formulating strategic managerial decisions that foster sustained growth. In this light, we extract medical service-related information from physicians' response behavioral data on a prominent healthcare platform, and empirically investigate the factors affecting physicians' online performance from a competitive perspective as well as the gender differences in these effects. The results indicate that physicians' responses significantly impact their online performance, revealing a competitive relationship between physicians and their colleagues in the same department. Specifically, a fast response time and informative responses are positively correlated with the focal physician's performance, whereas colleagues' informative responses negatively impact the focal physician's performance, and this relationship is mediated by the focal physician's response informativeness. Nevertheless, there is no significant correlation between colleagues' response time and the focal physician's performance. The results also unveil that gender moderates the effect of response informativeness on the focal physician's performance. Specifically, colleagues' response informativeness has a more significant impact on male physicians' performance than on female physicians' performance, suggesting a greater propensity for competition among male physicians. Our findings could offer decision support for enhancing physician performance and healthcare platform management.

Although algorithms are increasingly used to support professional tasks and routine decision-making, their opaque nature invites resistance and results in suboptimal use of their advice. Scholars argue for transparency to enhance the acceptability of algorithmic advice. However, current research is limited in understanding how improved transparency enhances the use of algorithmic advice, such as the differences among various aspects of transparency and the underlying mechanism. In this paper, we investigate whether and how different aspects of algorithmic transparency (performance, process, and purpose) enhance the use of algorithmic advice. Drawing on the knowledge-based trust perspective, we examine the mediating roles of trusting beliefs in the relationships between transparency and the use of algorithmic advice. Using the “judge-advisor system” paradigm, we conduct a 2 × 2 × 2 experiment to manipulate the three aspects of transparency and examine their effects on the use of algorithmic advice. We find that performance and process transparency promote the use of algorithmic advice. However, the effect of process transparency gets attenuated when purpose transparency is high. Purpose transparency is only useful when process transparency is low. We also find that while all three aspects of transparency facilitate different trusting beliefs, only competence belief significantly promotes the use of algorithmic advice. It also fully mediates the facilitating effects of performance and process transparency. This study contributes to the emerging research on algorithmic decision support by empirically investigating the effects of transparency on the use of algorithmic advice and identifying the underlying mechanism. The findings also provide practical guidance on how to promote the acceptance of algorithmic advice that is valuable to both individual users and practitioners.

Cyber threat intelligence (CTI) researchers strive to uncover collaborations and emerging techniques within hacker networks. This study proposes an empirical approach to detect communities within hacker forums for CTI purposes. Eighteen algorithms are systematically evaluated, including state-of-the-art and benchmark methods for identifying overlapping and disjoint groups. Using discussions from five prominent English hacker forums, a comparative analysis examines the influence of the algorithms’ theoretical foundations on community detection. Since ground truths are unattainable for such networks, the study utilizes a multi-metric strategy, incorporating modularity, coverage, performance, and a newly introduced quality measure, Triplet Hub Potential, which quantifies the presence of influential hubs. The findings reveal that while modularity optimization algorithms such as Leiden and Louvain deliver consistent results, neighbor-based expanding techniques tend to provide superior performance. In particular, the Expansion algorithm stood out by uncovering granular hierarchical community structures. The ability to investigate these intimacies is helpful for CTI researchers. Ultimately, we suggest an approach to investigate hacker forums using community detection methods and encourage the future development of algorithms tailored to expose nuances within hacker networks.

Developing new products by multiple stakeholders is inclined to project delays and even failures due to complex risk propagation, calling for accurate predictions of varying risk states and stakeholders' potential response actions. This study proposes an adaptive simulation-based decision support approach, starting with an adaptive simulation model capable of generating future intervention actions on risk propagation by mimicking stakeholders' risk response decisions. Accordingly, the approach tailors a genetic algorithm to solve the proposed simulation optimization problem and produce a combination of response actions that optimally block risk propagation at the current stage. To control dynamic propagations timely, this approach allows managers to adjust risk control resources in line with the latest risk states, and become accessible to managers by developing a graphical user interface. The application to a real project enables the validation of the usefulness and practicality of the approach.

This paper examines video games, a form of digital innovation, and seeks to predict a successful game based on the composition of game development team members. Team composition is measured with observable features generated from a graph network based on development team information derived from individual team member work on previous games. Features include network features, such as team member closeness, success percentile, and failure percentile, and non-network features, such as the number of games published prior by the studio. We propose a novel framework using these features to predict the chance of success for new games with an accuracy higher than 92%. Further, we investigate important features for prediction and provide model interpretability for practical implementations. We then build a decision support tool that allows video game producers, and associated stakeholders such as investors, to understand how the predictive model decides, predicts, and performs its recommendations. The findings have implications for those seeking to proactively impact digital product performance through graph network-generated features of team composition, where features are directly observable, as opposed to features that are more challenging to observe, such as personalities.

Information on firms' sustainability often partly resides in unstructured data published, for instance, in annual reports, news, and transcripts of earnings calls. In recent years, researchers and practitioners have started to extract information from these data sources using a broad range of natural language processing (NLP) methods. While there is much to be gained from these endeavors, studies that employ these methods rarely reflect upon the validity and quality of the chosen method—that is, how adequately NLP captures the sustainability information from text. This practice is problematic, as different NLP techniques lead to different results regarding the extraction of information. Hence, the choice of method may affect the outcome of the application and thus the inferences that users draw from their results. In this study, we examine how different types of NLP methods influence the validity and quality of extracted information. In particular, we compare four primary methods, namely (1) dictionary-based techniques, (2) topic modeling approaches, (3) word embeddings, and (4) large language models such as BERT and ChatGPT, and evaluate them on 75,000 manually labeled sentences from 10-K annual reports that serve as the ground truth. Our results show that dictionaries have a large variation in quality, topic models outperform other approaches that do not rely on large language models, and large language models show the strongest performance. In large language models, individual fine-tuning remains crucial. One-shot approaches (i.e., ChatGPT) have lately surpassed earlier approaches when using well-designed prompts and the most recent models.

Biometric authentication has become ubiquitous and profoundly impacts decision-making for both individuals and firms. Despite its extensive implementation, there is a discernible knowledge gap in understanding the nuanced influence of biometric characteristics on user security behaviors. To advance this line of research, we embrace the controversial information technology framework to delve into the juxtaposed nature of biometric characteristics, wherein they concurrently yield benefits and raise concerns that manifest in opposing effects on users' switching behavior. Adopting a sequential mixed methods approach, we first conducted semi-structured interviews that uncovered three key biometric characteristics and identified two benefits and two concerns associated with them. A follow-up survey was conducted to explore the interplay between each identified construct. The results emphasize the pivotal role of biometric characteristics in shaping user security behavior. Our research contributes to theoretical understanding by scrutinizing user behaviors vis-à-vis biometric authentication through a controversial IT perspective.

Competitive intelligence is vital for enterprises to survive in the market. Recently, online reviews have gained popularity among enterprises and researchers as a means to acquire timely and precise competitive insights. However, extant studies overlook the evolution of competitive information because they do not account for the variation of online reviews and products. In this research, we propose a method for dynamic competitive analysis by concentrating on the changes in products and online reviews. First, products and their related online reviews are analyzed via Dynamic Topic Model to derive product features mentioned in different slices. Second, we use sentiment analysis to estimate product performance and transfer the results into a product competitive relation network. Third, we implement competitive analysis from the perspectives of products and markets based on competitiveness propagation. By tracking the evolution of competitive relations among products, we discover competitors and glean more competitive insights. Lastly, a case study of laptops is used for validation. Experimental results indicate that our method is effective in capturing evolving and potential competitive relations among products.