Predicting aspect-based sentiment using deep learning and information visualization: The impact of COVID-19 on the airline industry

2022 | Information & Management | Citations: 0

Authors: Chang, Yung-Chun; Ku, Chih-Hao; Nguyen, Duy-Duc Le

Abstract: This study investigates customer satisfaction through aspect-level sentiment ana ... Expand

Abstract: This study investigates customer satisfaction through aspect-level sentiment analysis and visual analytics. We collected and examined the flight reviews on TripAdvisor from January 2016 to August 2020 to gauge the impact of COVID-19 on passenger travel sentiment in several aspects. Till now, information systems, management, and tourism research have paid little attention to the use of deep learning and word embedding techniques, such as bidirectional encoder representations from transformers, especially for aspect-level sentiment analysis. This paper aims to identify perceived aspect-based sentiments and predict unrated sentiments for various categories to address this research gap. Ultimately, this study complements existing sentiment analysis methods and extends the use of data-driven and visual analytics approaches to better understand customer satisfaction in the airline industry and within the context of the COVID-19. Our proposed method outperforms baseline comparisons and therefore contributes to the theoretical and managerial literature. Collapse

Topics: pandemic customer satisfaction visual analytics cloud computing data visualization

Methods: BERT deep learning natural language processing exploratory data analysis sentiment analysis

Theories: construal level theory appraisal theory

Factors influencing the acceptance of healthcare information technologies: A meta-analysis

2022 | Information & Management | Citations: 0

Authors: Chong, Alain Yee Loong; Blut, Markus; Zheng, Shuning

Abstract: Healthcare information technologies (HIT) can address several challenges faced b ... Expand

Abstract: Healthcare information technologies (HIT) can address several challenges faced by healthcare systems. To benefit from the advantages HIT offer, users must first accept them. This meta-analysis synthesizes previous research on HIT acceptance. It uses data from 214 independent samples reported in 193 articles and 83,619 technology users from 33 countries. The study contributes to the HIT literature by (1) synthesizing the empirical findings on technology acceptance factors and combining them in a comprehensive model, (2) testing the mediating mechanisms of health technology acceptance, and (3) examining contextual differences. The study finds that HIT acceptance depends on various predictors proposed by the technology acceptance model and the unified theory of acceptance and use of technology. These factors displayed strong indirect effects through effort expectancy, perceptions of the technology, performance expectancy, and attitudes toward using HIT. Studies overlooking these effects may underestimate the importance of various acceptance factors. Finally, the results suggest that technology acceptance varies across healthcare technologies (remote information systems [IS], wearables), users (staff/patients, age, voluntariness, experience), and locations (hospitals, healthcare systems, life expectancy in country). We also provide IS managers with guidance for improving technology acceptance in the healthcare industry to ensure efficient, high-quality services. Collapse

Topics: smart wearable device health information system technology adoption usage intention mobile system

Methods: meta analysis structural equation modeling literature sample literature study conceptual modelling

Theories: technology acceptance model unified theory of acceptance and use of technology TAM2 TAM3 extended unified theory of acceptance and use of technology

Working Hybrid at Universities: Old, yet New Practice?

2022 | International Conference on Information Systems | Citations: 0

Authors: Feiten Haubrich, Gislene; Hafermalz, Ella

Abstract: This study focuses on hybrid working in the university sector. The diversity of ... Expand

Abstract: This study focuses on hybrid working in the university sector. The diversity of the roles performed in universities provides a unique opportunity to study how the 'where and when' of knowledge work is evolving in the post-pandemic era. Our research aims to understand how hybrid working is 1) planned and 2) practiced in a university context. Through an analysis of ten university policies, we find that the ambition for hybrid working reveals several contradictions. Further, technology is mostly backgrounded in discussions of this new way of working. These preliminary findings challenge us consider how the initiatives predicted in a hybrid working policy take place in practice, at a large Dutch university. Based on interviews and systematic observations, we aim to enrich discussions of hybrid working with an open stance towards how and what role technology plays in the new era of work in the university context. Collapse

Topics: pandemic laptop computer productivity IT supported collaboration organizational context

Methods: field study qualitative interview design artifact laboratory experiment personal interview

A Boundary Spanning Perspective of Practical Impact: The Case of IS Practitioner Doctorates

2022 | International Conference on Information Systems | Citations: 0

Authors: McCarthy, Stephen; Scholta, Hendrik; Hausvik, Geir Inge; Busch, Peter André

Abstract: IS research often seeks to deliver practical impact, in addition to the traditi ... Expand

Abstract: IS research often seeks to deliver practical impact, in addition to the traditional requirement for theoretical contribution. While an admirable goal, it is nevertheless a challenging prospect, as key questions remain around how to best facilitate a relationship between IS academic and practitioner communities. To explore this question, our paper investigates boundary spanning by 'practitioner doctorates' -PhD students with professional experience who seek to span the fields of academia and practice during their research. Drawing on in-depth interviews with practitioner doctorates, our findings point towards several factors for practical impact such as researcher legitimacy, expectation management, and adapting to changes in industry requirements. In doing so, we contribute towards an understanding of engaged scholarship in IS and take steps towards addressing the dearth of research on doctoral studies in the IS field to date. Collapse

Topics: boundary spanning systems implementation digital transformation technology adoption decision making

Methods: qualitative interview personal interview qualitative content analysis theoretical contribution case study

Co-Governance in Digital Transformation Initiatives: The Roles of Digital Culture and Employee Experience

2021 | HICSS | Citations: 0

Authors: Abhari, Kaveh; Ostroff, Chloe; Barcellos, Bailey; Williams, David

Abstract: Digital transformation (DT) is increasingly fundamental for organizations to not ... Expand

Abstract: Digital transformation (DT) is increasingly fundamental for organizations to not only implement, but thoroughly understand and dictate. Recent studies suggest that DT is not limited to the process of implementing digital technology to enhance business performance; it is the process of harmonizing organizational goals, values, and culture with employees by the mean of digital technologies. Therefore, it is critical to understand DT and determine its success from the perspective of the employee. To further understand the role of employees in DT, this paper theorizes and validates the relationships between digital culture, employee experience with DT, and DT co-governance. The findings guide theoretical and practical development in the field. Collapse

Topics: IT governance organizational culture business process management information system use social network

Methods: survey theory development partial least squares regression psychometrics descriptive statistic

Social enterprises in electronic markets: web localization or standardization

2021 | Electronic Markets | Citations: 1

Authors: Benmamoun, Mamoun; Alhor, Hadi; Ascencio, Christine; Sim, Woojong

Abstract: The question of how social enterprises (SEs)—organizations that work to solve so ... Expand

Abstract: The question of how social enterprises (SEs)—organizations that work to solve social issues through innovative ideas and strategies—utilize the web to internationalize has received little attention in electronic business literature. We address this issue by examining whether SEs pursue a web globalization strategy that is standardized across national markets or localized and culturally customized to individual locales. Accepted knowledge about e-commerce corporations predicts that these firms would be better off with web localization. However, an investigation using content analysis and two case studies complemented with face-to-face and digital interviews reveals that SEs differ inherently from for-profit businesses in web globalization. Specifically, this paper shows that SEs adopt a web strategy high on standardization to mobilize, educate, advocate, and recruit would-be donors, influencers, and volunteers. Such a strategy allows SEs to build legitimacy and reinforce their global brand. Our empirical findings show the uniqueness of SEs’ digital context from profit-oriented e-commerce corporations and the need for a suitable theoretical framework. Collapse

Topics: website electronic market electronic business entrepreneurship electronic mail

Methods: case study qualitative content analysis qualitative interview qualitative coding personal interview

Theories: internationalization theory

Classification of Real-World Microgrids Based on a Morphological Analysis

2021 | Americas Conference on Information Systems | Citations: 0

Authors: Gerlach, Jana; Eckhoff, Sarah; Werth, Oliver; Kraschewski, Tobias; Brauner, Tim; Breitner, Michael H.

Abstract: Microgrids integrate distributed energy resources into an energy network reliab ... Expand

Abstract: Microgrids integrate distributed energy resources into an energy network reliably and efficiently. However, research of real-world examples at the international level is limited. We conduct a morphological analysis for microgrid design options to examine the status quo of academic literature. We identify 18 dimensions with 60 characteristics divided into the five layers governance, business, intelligence, communication, and physical. Subsequently, we classify 30 real-world microgrids with diverse types and locations using our morphological box. Our analysis reveals future research requirements regarding social aspects, business models, critical success factors, and maturity levels. We provide a framework supporting decision-makers to identify microgrid design options and promote socially, economically, and environmentally sustainable, resilient, and decentralized energy supply. Collapse

Topics: business model Google+ critical success factor database system website

Methods: literature study literature sample mixed method case study

Is Hidden Safe? Location Protection against Machine-Learning Prediction Attacks in Social Networks

2021 | Management Information Systems Quarterly | Citations: 0

Authors: Han, Xiao; Wang, Leye; Fan, Weiguo

Abstract: User privacy protection is a vital issue of concern for online social networks ... Expand

Abstract: User privacy protection is a vital issue of concern for online social networks (OSNs). Even though users often intentionally hide their private information in OSNs, since adversaries may conduct prediction attacks to predict hidden information using advanced machine learning techniques, private information that users intend to hide may still be at risk of being exposed. Taking the current city listed on Facebook profiles as a case, we propose a solution that estimates and manages the exposure risk of users' hidden information. First, we simulate an aggressive prediction attack using advanced state-of-the-art machine learning algorithms by proposing a new current city prediction framework that integrates location indications based on various types of information exposed by users, including demographic attributes, behaviors, and relationships. Second, we study prediction attack results to model patterns of prediction correctness (as correct predictions lead to information exposures) and construct an exposure risk estimator. The proposed exposure risk estimator has the ability not only to notify users of exposure risks related to their hidden current city but can also help users mitigate exposure risks by overhauling and selecting countermeasures. Moreover, our exposure risk estimator can improve the privacy management of OSNs by facilitating empirical studies on the exposure risks of OSN users as a group. Taking the current city as a case, this work offers insight on how to protect other types of private information against machine-learning prediction attacks and reveals several important implications for both practice management and future research. Collapse

Topics: privacy Facebook risk management geographic information system information exchange

Methods: machine learning computational algorithm experiment support vector machine digital trace data

Identifying Sentiment Influences Provoked by Context Factors – Results from a Data Analytics Procedure Performed on Tweets

2021 | HICSS | Citations: 0

Authors: Konadl, Daniel; Wörner, Janik; Leist, Susanne

Abstract: Context factors have lasting impacts on people’s sentiments. Exploring impacts t ... Expand

Abstract: Context factors have lasting impacts on people’s sentiments. Exploring impacts that different contexts have on sentiments can be crucial for managing the increasing number of communications companies nowadays maintain with customers via social media channels. To help companies prevent impacts of negative word of mouth, we provide an overview about sentiment-influential contexts for tweets as one kind of social media texts previously discussed within the literature. We collected an overall amount of 358.923.210 tweets and performed analysis to uncover the effects of continents, mobile devices’ operating systems (OS) and the combination of both on sentiments expressed within tweets. Our results show remarkable differences for tweets originating from North America and Apple devices, which turned out to be the tweets with the lowest sentiments compared to the other continents and the mobile OS Android. Collapse

Topics: Twitter mobile system Android social media word of mouth

Methods: Student's t-test sentiment analysis literature study natural language processing

The Pragmatic School of Thought in Open Science: A Case Study of Multi- stakeholder Participation in Shaping the Future of Internet Governance

2021 | HICSS | Citations: 0

Authors: McCarthy, Stephen; Mahony, Carolanne; Rowan, Wendy; Tran-Karcher, Huy; Tran, Huy; Potet, Manon

Abstract: The internet is a disruptive technology that continues to define our modern worl ... Expand

Abstract: The internet is a disruptive technology that continues to define our modern world. However, numerous ethical challenges remain for internet governance going forward, e.g. surveillance capitalism, terrorism and radicalisation. The ‘pragmatic’ school of thought in open science advocates for collaboration between diverse stakeholder groups (e.g. citizens, academics, practitioners, policymakers) to ensure an informed, and positive imprint for change. However, our understanding of how open science can be used for assimilating knowledge on complex socio-political issues remains nascent. To address this gap, we present findings from ‘We, the Internet’, a global consultation project which utilised open science practices such as stakeholder-led evaluations and open access publications to engage stakeholders in dialogue around the future of internet governance. Our findings discuss emergent themes on the future of internet governance, and highlight the potential of open science to mobilise groups and combat public scepticism in policy-making. Collapse

Topics: IT governance internet technology open access citizen science information system use

Methods: case study survey