This study aims to address quality concerns in mobile Extended Reality (XR) apps by developing tools to classify user reviews from the Google Play Store. Our first major contribution is a holistic approach to coding thousands of reviews, offering a unified framework to assess feature concerns and feature suggestions, thereby enhancing the understanding of user expectations. We also introduce a novel toolset for extracting key terms related to these concerns and suggestions, facilitating rapid analysis and improvement of app functionalities. To validate our approach, we compare the performance of this automated coding method with sentiment analysis, deep learning-based BERT, and several large language models (LLMs). Additionally, our study extends the understanding of user satisfaction by analyzing quantitative indicators such as app ratings and review helpfulness, offering new perspectives on user perceptions of XR app quality. This research has discovered vital areas for feature improvement without implementing complex infrastructure and using advanced technical expertise, enabling app firms to better target their efforts in improving XR app experiences.

Many firms face challenges in monitoring their products for evidence of safety hazards, which can have profoundly negative effects both on consumers and on firms’ financial standings. Monitoring online reviews has the potential to unveil vital safety insights, but the volume of the data poses practical challenge. A popular recent approach to safety surveillance has involved utilizing “smoke terms” customized to the safety-related language in each product category. In this work, we suggest two new methods for deriving smoke terms: piecewise smoke terms and sum of ranks smoke terms. In addition, we address the question of whether smoke terms can be trained to perform well across product categories. We find that our new smoke term methods show improvement in performance upon traditional approaches, such as sentiment analysis, for detecting mentions of safety hazards. We find that piecewise smoke terms in particular performed well across all multiple product categories. We also found that while it is difficult for a cross-category technique to match the performance of a category-specific technique, our novel set of cross-category smoke terms performed with great promise and offer substantial improvement upon sentiment analysis.

In recent years, online reviews have offered a rich new medium for consumers to express their opinions and feedback. Product designers frequently aim to consider consumer preferences in their work, but many firms are unsure of how best to harness this online feedback given that textual data is both unstructured and voluminous. In this study, we use text mining tools to propose a method for rapid prioritization of online reviews, differentiating the reviews pertaining to innovation opportunities that are most useful for firms. We draw from the innovation and entrepreneurship literature and provide an empirical basis for the widely accepted attribute mapping framework, which delineates between desirable product attributes that firms may want to capitalize upon and undesirable attributes that they may need to remedy. Based on a large sample of reviews in the countertop appliances industry, we demonstrate the performance of our technique, which offers statistically significant improvements relative to existing methods. We validate the usefulness of our technique by asking senior managers at a large manufacturing firm to rate a selection of online reviews, and we show that the selected attribute types are more useful than alternative reviews. Our results offer insight in how firms may use online reviews to harness vital consumer feedback.

Organizations have recognized that engaging with customers on social media is now a critical element of their public relations strategy, but many struggle to find the optimal way to use these new tools. This study examined the success factors of online posts from organizations in evoking a high level of response from group members. Specifically, appeals to community belongingness were discovered to be effective at motivating user engagement on social media. The impact of community‐focused language on garnering likes, comments, and shares was assessed using a methodology that combined text analytics, supervised machine learning, and multilevel regression. In 51,760 postings from 100 academic libraries, it was found that general community appeals increase likes by 60%, comments by 130%, and shares by 39%, whereas specific intracommunity, locally meaningful references increase likes by 16% and shares by 12%. These findings have both theoretical implications, as they constitute a large‐scale, empirical confirmation of the belongingness hypothesis from social psychology, and practical implications, as they suggest best practices for maintaining an organizational online presence.

Nonprofit organizations are pivotal actors in society, working in various sectors attempting to fill gaps between government services and the market. This interdisciplinary research compared nonprofit websites in Thailand, a developing country, to nonprofit websites in the United States to determine progression through an adapted e-business stage model. A manual decoding process of websites and a web crawling analysis software, Sitebeam, were used to determine the nonprofits’ e-progression. Thai and U.S. nonprofits did not differ significantly from each other in terms of presence building content or interactivity, but the U.S. websites were more advanced in e-transactions than the Thai websites (55% vs. 37%). The acceptance of e-transactions was also found to be highly associated with greater online presence and interactivity, suggesting that there is significant room for advancement among organizations that do not accept e-transactions both in Thailand and the United States

Many documents containing information about intelligence and security issues are available both in printed and electronic formats. In this research, we built an experimental system to extract intelligence and security information from electronic documents. Our system, CAINES, is based on a knowledge engineering approach and relies on sublanguage analysis techniques. CAINES performs syntactic and semantic analysis and uses lexicons of various categories of terms. The system is able to extract certain types of information from reports on terrorist incidents posted by the National Counterterrorism Center (NCTC), such as what happened and the results of the incidents.

This paper proposes a framework for cost-sensitive classification under a generalized cost function. By combining decision trees with sequential binary programming, we can handle unequal misclassification costs, constrained classification, and complex objective functions that other methods cannot. Our approach has two main contributions. First, it provides a new method for cost-sensitive classification that outperforms a traditional, accuracy-based method and some current cost-sensitive approaches. Second, and more important, our approach can handle a generalized cost function, instead of the simpler misclassification cost matrix to which other approaches are limited.