Increasingly, machine learning methods have been applied to aid in diagnosis with good results. However, some complex models can confuse physicians because they are difficult to understand, while data differences across diagnostic tasks and institutions can cause model performance fluctuations. To address this challenge, we combined the Deep Ensemble Model (DEM) and tree-structured Parzen Estimator (TPE) and proposed an adaptive deep ensemble learning method (TPE-DEM) for dynamic evolving diagnostic task scenarios. Different from previous research that focuses on achieving better performance with a fixed structure model, our proposed model uses TPE to efficiently aggregate simple models more easily understood by physicians and require less training data. In addition, our proposed model can choose the optimal number of layers for the model and the type and number of basic learners to achieve the best performance in different diagnostic task scenarios based on the data distribution and characteristics of the current diagnostic task. We tested our model on one dataset constructed with a partner hospital and five UCI public datasets with different characteristics and volumes based on various diagnostic tasks. Our performance evaluation results show that our proposed model outperforms other baseline models on different datasets. Our study provides a novel approach for simple and understandable machine learning models in tasks with variable datasets and feature sets, and the findings have important implications for the application of machine learning models in computer-aided diagnosis.

Purpose With the popularity of the internet, access to health-related information has become more convenient. However, the easy acquisition of e-health information could lead to unfavorable consequences, such as health anxiety. The purpose of this paper is to explore a set of important influencing factors that lead to health anxiety. Design/methodology/approach Based on the stimulus–organism–response (S-O-R) framework, we propose a theoretical model of health anxiety, with metacognitive beliefs and catastrophic misinterpretation as the mediators between stimulus factors and health anxiety. Using 218 self-reported data points, the authors empirically examine the research model and hypotheses. Findings The study results show that anxiety sensitivity positively affects metacognitive beliefs. The severity of physical symptoms has a significant positive impact on catastrophic misinterpretation. Metacognitive beliefs and catastrophic misinterpretation have significant positive impacts on health anxiety. Originality/value Based on the S-O-R model, this paper develops a comprehensive model to explain health anxiety and verifies the model using firsthand data.

Observing the positive aspects of others' lives on social media (SM) can bring about envy among users. Drawing from social comparison and technology acceptance theories, this study develops a research model to explain how envy occurs and impacts SM users' behavior. In this work, we conducted two studies across three different SM settings to investigate two types of envy, benign and malicious envy. The results show that malicious envy is negatively related to the dependent variable of SM use intention while benign envy facilitates it. The findings provide many valuable contributions to both information systems (IS) academia and industry. This study identifies the unique SM factors intertwining with envy. Moreover, this work helps SM users and practitioners be aware of the potential envy issue on SM so they can take effective actions to enhance SM use.

Background: With the continuous development of the internet and the explosive growth in data, big data technology has emerged. With its ongoing development and application, cloud computing technology provides better data storage and analysis. The development of cloud health care provides a more convenient and effective solution for health. Studying the evolution of knowledge and research hotspots in the field of cloud health care is increasingly important for medical informatics. Scholars in the medical informatics community need to understand the extent of the evolution of and possible trends in cloud health care research to inform their future research. Objective: Drawing on the cloud health care literature, this study aimed to describe the development and evolution of research themes in cloud health care through a knowledge map and common word analysis. Methods: A total of 2878 articles about cloud health care was retrieved from the Web of Science database. We used cybermetrics to analyze and visualize the keywords in these articles. We created a knowledge map to show the evolution of cloud health care research. We used co-word analysis to identify the hotspots and their evolution in cloud health care research. Results: The evolution and development of cloud health care services are described. In 2007-2009 (Phase I), most scholars used cloud computing in the medical field mainly to reduce costs, and grid computing and cloud computing were the primary technologies. In 2010-2012 (Phase II), the security of cloud systems became of interest to scholars. In 2013-2015 (Phase III), medical informatization enabled big data for health services. In 2016-2017 (Phase IV), machine learning and mobile technologies were introduced to the medical field. Conclusions: Cloud health care research has been rapidly developing worldwide, and technologies used in cloud health research are simultaneously diverging and becoming smarter. Cloud-based mobile health, cloud-based smart health, and the security of cloud health data and systems are three possible trends in the future development of the cloud health care field.

Online petitions have become a powerful tool used by the public to affect change in society. Despite the increasing popularity of these petitions, it remains unclear how the public consumes and interprets their content and how this helps the creators of online petitions achieve their goals. This study investigates how linguistic factors present in online petition texts influence petition success. Specifically, drawing upon the dual-process theory of persuasion and the moral persuasion literature, this study examines cognitive, emotional, and moral linguistic factors in petition texts and identifies how they contribute to the success or failure of online petitions. The results, which are based on an analysis of 45,377 petitions from Change.org , show that petitions containing positive emotions and enlightening information are more likely to succeed. Contrary to popular belief, petitions containing heavy cognitive reasoning and those emphasizing moral judgment are less likely to succeed. This study also exemplifies the use of an analytical approach for examining crowd-sourced content involving online political phenomena related to policy-making, governance, political campaigns, and large social causes.

With the continuous development of the Internet and the explosive growth of data, big data technology emerged. The development and application of cloud computing technology provides better storage and analysis of data. The development of cloud healthcare provides a more convenient and effective solution for people’s health. To study the knowledge evolution in the field of cloud healthcare and the research hot topics is becoming one of important issues in medical informatics area. The scholars in medical informatics community need to understand the panorama of the evolution of cloud healthcare research, as well as possible trends in cloud healthcare field as important reference for their future research work. Drawing on the cloud healthcare literature, this paper aims at revealing the development and evolution of research themes in cloud healthcare through knowledge map and common word analysis. A total of 2878 articles in the cloud healthcare literature were retrieved from the Web of Science database. We used cybermetrics to analyze and visualize the keywords in these articles. In particular, we create a knowledge map to show the evolution of cloud healthcare research. We use co-word analysis to reveal the hot topics in cloud healthcare research and their evolution process. The evolution and development of cloud healthcare services are shown. And From 2007 to 2009 (Phase I), most scholars applied cloud computing to the medical field mainly to reduce cost; and the technologies used are primarily grid computing and cloud computing. During 2010-2012 (Phase II), scholars began to pay attention to the security of cloud systems. During 2013-2015 (Phase III), the medical informatization created the big data for health services. During 2016-2017 (Phase IV), machine learning and mobile technologies are introduced into the medical field. The research of cloud healthcare has been vigorously developing worldwide, and technologies used in cloud health research are becoming diverged and smart simultaneously. Cloud-based mobile health, cloud-based smart health, as well as the security of cloud health data and systems will be three possible trends in the future development of the cloud healthcare field.

With the rise of artificial intelligence, case-based health knowledge management systems (CBHKS) have been widely adopted in hospitals. CBHKS are data-driven intelligent platforms that integrate latest technologies, such as artificial intelligence and cloud computing. As an integral part of smart hospitals, CBHKS can support decision processes at different levels in hospitals. However, researchers have not yet clearly addressed how CBHBKS improves hospital management outcomes. Based on group effectiveness and leadership performance-maintenance theories, we develop a conceptual model to explain the role of CBHKS in hospital management. To test the research hypotheses in the conceptual model, we collected survey data from 214 doctors, and performed data analysis using partial least squares (PLS)-based structural equation modeling. The empirical testing results show that the CBHKS implementation significantly and positively influences group performance, group members’ satisfaction, group learning, and external satisfaction; and group members’ satisfaction and external satisfaction significantly and positively affect management performance and maintenance.

Massive social media data present businesses with an immense opportunity to extract useful insights. However, social media messages typically consist of both facts and opinions, posing a challenge to analytics applications that focus more on either facts and opinions. Distinguishing facts and opinionss may significantly improve subsequent analytics tasks. In this study, we propose a deep learning-based algorithm that automatically separates facts from opinions in Twitter messages. The algorithm outperformed multiple popular baselines in an experiment we conducted. We further applied the proposed algorithm to track customer complaints and found that it indeed benefits subsequent analytics applications.

Hofstede's work on national culture has been extensively used in cross-national studies in the information systems discipline. In particular, many cross-national cultural researchers have used Hofstede's cultural index. This study argues that espoused national cultural values should be measured when the unit of analysis of the cross-national cultural study is the individual. This study reviews cross-national studies published in eight IS journals and examines both cross-national studies and cross-national cultural studies. After that, this work provides rationales of why espoused national cultural values should be measured. Finally, we conclude that espoused national culture is more appropriate for individual behavior research.

Envy is receiving more and more attention from both the IT industry and academia. IT researchers usually rely on subjective, survey-based methods to study envy. In the current paper, we investigated envy-related messages from Twitter through text analytics. We found that online envy was significantly associated with social media use patterns: users who "favorite" more and post less are more likely to express malicious envy while users who "favorite" less and post more are more likely to express benign envy. Moreover, users' influence moderates the degree of the two types of envy. When a user is a profound influencer, the user would be more likely to express a lower level of malicious envy, and would also be more likely to express a higher level of benign envy. The findings provide a number of theoretical contributions and practical implications.

THREE ESSAYS ON HUMAN BEHAVIORS IN SOCIAL MEDIA