Decision Support Systems最新文献

Electronic health records (EHRs) are a valuable source of information that can aid in understanding a patient’s health condition and making informed healthcare decisions. However, modelling longitudinal EHRs with heterogeneous information is a challenging task. Although recurrent neural networks (RNNs) are frequently utilized in artificial intelligence (AI) models for capturing longitudinal data, their explanatory capabilities are limited. Predictive clustering stands as the most recent advancement within this domain, offering interpretable indications at the cluster level for predicting disease risk. Nonetheless, the challenge of determining the optimal number of clusters has put a brake on the widespread application of predictive clustering for disease risk prediction. In this paper, we introduce a novel non-parametric predictive clustering-based risk prediction model that integrates the Dirichlet Process Mixture Model (DPMM) with predictive clustering via neural networks. To enhance the model’s interpretability, we integrate attention mechanisms that enable the capture of local-level evidence in addition to the cluster-level evidence provided by predictive clustering. The outcome of this research is the development of a multi-level explainable artificial intelligence (AI) model. We evaluated the proposed model on two real-world datasets and demonstrated its effectiveness in capturing longitudinal EHR information for disease risk prediction. Moreover, the model successfully produced interpretable evidence to bolster its predictions.

Customer retention management relies on advanced analytics for decision making. Decision makers in this area require methods that are capable of accurately predicting which customers are likely to churn and that allow to discover drivers of customer churn. As a result, customer churn prediction models are frequently evaluated based on both their predictive performance and their capacity to extract meaningful insights from the models. In this paper, we extend hybrid segmented models for customer churn prediction by incorporating powerful models that can capture non-linearities. To ensure the interpretability of such segmented hybrid models, we introduce a novel model-agnostic approach that extends SHAP. We extensively benchmark the proposed methods on 14 customer churn datasets on their predictive performance. The interpretability aspect of the new model-agnostic approach for interpreting hybrid segmented models is illustrated using a case study. Our contributions to decision making literature are threefold. First, we introduce new hybrid segmented models as powerful tools for decision makers to boost predictive performance. Second, we provide insights in the relative predictive performance by an extensive benchmarking study that compares the new hybrid segmented methods with their base models and existing hybrid models. Third, we propose a model-agnostic tool for segmented hybrid models that provide decision makers with a tool to gain insights for any hybrid segmented model and illustrate it on a case study. Although we focus on customer retention management in this study, this paper is also relevant for decision makers that rely on predictive modeling for other tasks.

Hepatitis carcinoma (HCC) accounts for the majority of liver cancer–related deaths globally. Cirrhosis often precedes HCC clinically in a strong, temporal relationship. Therefore, identifying cirrhosis patients at higher risk of HCC is crucial to physicians' clinical decision-making and patient management. Effective estimates of at-risk patients can facilitate timely therapeutic interventions and thereby enhance patient outcomes and well-being. We develop a novel, meta-path, attention-based deep learning method to identify at-risk cirrhosis patients. The proposed method integrates complex patient–medication interactions, essential patient–patient and medication–medication links, and the combined effects of medication and comorbidity to support downstream predictions. An empirical test of the proposed method's predictive utilities, relative to nine existing methods, uses a large sample of real-world cirrhosis patient data. The comparative results indicate that the proposed method can identify at-risk patients more effectively than all the benchmarks. The current research has important implications for clinical decision support and patient management, and it can facilitate patient self-management and treatment compliance too.

Employees' reviews, feedback, opinions, and experiences shared on crowdsourcing platforms are now widely used by human resource management researchers to analyze a firm's performance, management effectiveness, and culture. The analysis of firm ratings posted by employees on crowdsourcing platforms can not only provide timely feedback and insights into a firm's operations but also inspire managers to make better decisions to improve organizational performance. Based on economic and psychological theories, we conduct a comprehensive and item-by-item analysis of firm ratings on Glassdoor using panel vector autoregression to explore the interactive relationship between crowdsourced firm ratings and Total Factor Productivity (TFP), examining whether this relationship differs across industries. We find a circular interaction between firms' overall ratings and TFP. Additionally, we explore employees' perspectives on compensation and work-life balance. Our results indicate that compensation ratings negatively impact TFP, whereas work-life balance ratings are solely influenced by the lagged self. Finally, we observe that the interaction between Glassdoor firm ratings and TFP varies across industries. Our study suggests that decision makers of different industries should tailor motivation strategies to suit the specific needs of their workforce, allocating resources differently between compensation and work-life balance initiatives.

This paper presents a comprehensive framework for expert-augmented supervised feature selection, addressing pre-processing, in-processing, and post-processing aspects of Explainable Artificial Intelligence (XAI). As part of pre-processing XAI, we introduce the Probabilistic Solution Generator through the Information Fusion (PSGIF) algorithm, leveraging ensemble techniques to enhance the exploration and exploitation capabilities of a Genetic Algorithm (GA). Balancing explainability and prediction accuracy, we formulate two multi-objective optimization models that empower expert(s) to specify a maximum acceptable sacrifice percentage. This approach enhances explainability by reducing the number of selected features and prioritizing those considered more relevant from the domain expert's perspective. This contribution aligns with in-processing XAI, incorporating expert opinions into the feature selection process as a multi-objective problem. Traditional feature selection techniques lack the capability to efficiently search the solution space considering our explainability-focused objective function. To overcome this, we leverage the Genetic Algorithm (GA), a powerful metaheuristic algorithm, optimizing its parameters through Bayesian optimization. For post-processing XAI, we present the Posterior Ensemble Algorithm (PEA), estimating the predictive power of features. PEA enables a nuanced comparison between objective and subjective importance, identifying features as underrated, overrated, or appropriately rated. We evaluate the performance of our proposed GAs on 16 publicly available datasets, focusing on prediction accuracy in a single objective setting. Moreover, we test our multi-objective model on a classification dataset to show the applicability and effectiveness of our framework. Overall, this paper provides a holistic and nuanced approach to explainable feature selection, offering decision-makers a comprehensive understanding of feature importance.

Patient experience surveys have become a key source of evidence for supporting decision-making and continuous quality improvement within healthcare services. To harness free-text feedback collected as part of these surveys for additional insights, text analytics methods are increasingly employed when the data collected is not amenable to traditional qualitative analysis due to volume. However, while text analytics techniques offer good predictive capabilities, they have limited explanatory features often required in formal decision-making contexts, such as programme monitoring or evaluation. To overcome these limitations, this study integrates computational text and predictive modelling as part of a Computational Grounded Theory method to determine the effect of quality gaps in care dimensions and their prioritisation from free-text feedback. The feedback was collected as part of a national survey to support decisions on continuous improvement in Maternity Services in Ireland. Our approach enables (1) operationalising the service quality lexicon in the context of maternity care to explain the effect of quality gaps in care dimensions on overall satisfaction from free-text comments; and (2) extending the service quality lexicon with two organisational and political decision-making concepts: “Salience” and “Valence”, for prioritising perceived quality gaps. These methodological affordances enable the extension of service quality theory to explicitly support the prioritisation of improvement decisions which before now required additional decision frameworks. Results show that tangibles-, process-, and reliability-related care issues have the highest importance in our study context. We also find that hospital contexts partly determine the relative importance of gaps in care dimensions.

This study proposes a real-time Decision Support System (DSS) using machine learning to enhance proactive management of Human–Machine Interaction (HMI) in safety–critical digital control rooms. The DSS provides explainable predictions and recommendations regarding near-future automation usage, customized for the railway control room management, who supervise the operations of traffic controllers (TCs). In this setting, TCs decide on the spot whether to manually or automatically open signals to regulate railway traffic, a critical aspect of ensuring punctuality and safety. This time-setting specific HMI differs across TCs and is not yet supported by a data-driven tool. The proposed DSS includes agreement levels for predictions among different modeling paradigms: linear models, tree-based models, and deep neural networks. SHAP (SHapley Additive exPlanations) values are deployed to assess the agreement level in explainability between these different modeling paradigms. The prescriptions are based on the HMI of well-performing peers. We implement the DSS as proof of concept at the Belgian railway infrastructure company and report end-user feedback on the perception, the operational impact, and the inclusion of agreement levels.

Despite having potentially important implications, there has been little research on the relationship between the public's incidental emotion and the stock market. To that end, we construct a valence-based measure of incidental emotion using BERTweet's sentiment analysis and empirically investigate the association between collective incidental emotion toward the COVID-19 pandemic and the U.S. stock market. We employ multivariate time series autoregressive models to test the relationship between emotion polarity and stock market returns or trading volumes. The results reveal that societal sentiment toward the pandemic has a significant effect on the returns of the Dow Jones Industrial Average and S&P 500. In contrast, the macro-level emotion does not significantly affect the return for NASDAQ 100. The findings also suggest a significant association between incidental emotion and trading volumes. We conduct a battery of sensitivity tests that further support our conjecture. The study underscores the robust role of incidental emotion in investment decision-making, highlighting its significance as a distinctive feature that should be incorporated into financial decision support systems.

Sectorization problems refer to dividing a large set, area or network into smaller parts concerning one or more objectives. A decision support system (DSS) is a relevant tool for solving these problems, improving optimisation procedures, and finding feasible solutions more efficiently. This paper presents a new web-based Decision Support System for Sectorization (D3S). D3S is designed to solve sectorization problems in various areas, such as school and health districting,planning sales territories and maintenance operations zones, or political districting. Due to its generic design, D3S bridges the gap between sectorization problems and a state-of-the-art decision support tool. The paper aims to present the generic and technical attributes of D3S by providing detailed information regarding the problem-solution approach (based on Evolutionary Algorithms), objectives (most common in sectorization), constraints, structure and performance.

The emergence of enterprise social media (ESM) allows enterprises to develop employee improvisation ability for effective decision-making in various emergencies. However, it remains unclear how the use of ESM by employees affects their ability to improvise. Based on the job demands-resources model and Kahn's psychological conditions framework, this study constructs a theoretical model capturing two types of ESM usage—work-related and social-related—and examines their impact on employee improvisation ability. Through the analysis of 307 paired data collected from multi-wave and multi-source questionnaires using Smart-PLS software, the results show that both work-related and social-related ESM use can promote employees' psychological meaningfulness, availability, and safety, thus further stimulating employees' improvisation ability. ESM policies only significantly moderated the effects of work-related ESM use on the three psychological conditions of employees. Moreover, there are significant differences in the intensity of the influence of the two types of ESM uses on the psychological conditions of employees. This study not only enriches and promotes the existing research on ESM usage, psychological conditions, and employee improvisation ability but also helps enterprise management effectively guide employees to use ESM to promote their improvisation ability.