App Categories Research Articles

Classifying Mobile Applications Using Word Embeddings

Modern application stores enable developers to classify their apps by choosing from a set of generic categories, or genres, such as health, games, and music. These categories are typically static—new categories do not necessarily emerge over time to reflect innovations in the mobile software landscape. With thousands of apps classified under each category, locating apps that match a specific consumer interest can be a challenging task. To overcome this challenge, in this article, we propose an automated approach for classifying mobile apps into more focused categories of functionally related application domains. Our aim is to enhance apps visibility and discoverability. Specifically, we employ word embeddings to generate numeric semantic representations of app descriptions. These representations are then classified to generate more cohesive categories of apps. Our empirical investigation is conducted using a dataset of 600 apps, sampled from the Education, Health&Fitness, and Medical categories of the Apple App Store. The results show that our classification algorithms achieve their best performance when app descriptions are vectorized using GloVe, a count-based model of word embeddings. Our findings are further validated using a dataset of Sharing Economy apps and the results are evaluated by 12 human subjects. The results show that GloVe combined with Support Vector Machines can produce app classifications that are aligned to a large extent with human-generated classifications.

A Multidimensional Analysis of Released COVID-19 Location-Based Mobile Applications

The spread of coronavirus disease (COVID-19) has triggered a series of responses worldwide ranging from traveling restrictions and shelter-in-place orders to lockdowns, contact tracing, social distancing, and other mitigation measures. To assist with contact tracing and ensure the safety of citizens, a significant number of mobile applications has been developed, utilizing geospatial information and proximity sensing. We perform a thorough research on seven digital databases (Appbrain, e-Health Hub, GDPRhub, “fs0c131y”, News Sites, Appstore, and Google Play), identifying a total of 160 apps regarding COVID-19 related to our research questions. The aim of this research is to identify the main categories of apps and analyze their functions based on a proposed framework of by mapping aspects that affect their functionalities regarding Services, Technology, Societal & Business, and Legal aspects. As the world comes to the new normal, the utilization of these apps might become more essential for more mobile users and developers. The new encryption protocols that are established are also in favor of this argument. Future work can utilize our framework to further examine the development, design, and adoption of such mobile applications.

Packet-level prediction of mobile-app traffic using multitask Deep Learning

The prediction of network traffic characteristics helps in understanding this complex phenomenon and enables a number of practical applications, ranging from network planning and provisioning to management, with security implications as well. A significant corpus of work has so far focused on aggregated behavior, e.g., considering traffic volumes observed over a given time interval. Very limited attempts can instead be found tackling prediction at packet-level granularity. This much harder problem (whose solution extends trivially to the aggregated prediction) allows a finer-grained knowledge and wider possibilities of exploitation. The recent investigation and success of sophisticated Deep Learning algorithms is now providing mature tools to face this challenging but promising goal.In this work, we investigate and specialize a set of architectures selected among Convolutional, Recurrent, and Composite Neural Networks, to predict mobile-app traffic at the finest (packet-level) granularity. We discuss and experimentally evaluate the prediction effectiveness of the provided approaches also assessing the benefits of a number of design choices such as memory size or multi-modality, investigating performance trends at packet level focusing on the head and the tail of biflows. We compare the results with both Markovian and classic Machine Learning approaches, showing increased performance with respect to state-of-the-art predictors (high-order Markov chains and Random Forest Regressor). For the sake of reproducibility and relevance to modern traffic, all evaluations are conducted leveraging two real human-generated mobile traffic datasets including different categories of mobile apps. The experimental results witness remarkable variability in prediction performance among different apps categories. The work also provides valuable analysis results and tools to compare different predictors and strike the best balance among the performance measures.

When to Be Agile: Ratings and Version Updates in Mobile Apps

Lean and agile models of product development organize the flexible capacity to rapidly update individual products in response to customer feedback. Although agile operations have been adopted across numerous industries, neither the benefits nor the factors explaining when firms choose to become agile are validated and understood. We study these questions using data on the development of mobile apps, which occurs through the dynamic release of new versions into the mobile app marketplace, and the apps’ customer ratings. We develop a structural model estimating the dependence of product versioning on (a) market feedback in the form of customer ratings against (b) project and work-based considerations, such as development timelines, scale economies, and operational constraints. In contrast to when they actually benefit from operational agility, firms become agile when launching riskier products (in terms of uncertainty in initial customer reception) and less agile when they are able to exploit scale economies from coordinating development over a portfolio of apps. Agile operations increase firm payoffs by margins of 20% to 80%, and interestingly, partial agility is often sufficient to capture the bulk of these returns. Finally, turning to a question of marketplace design, we study how the mobile app marketplace should design the display of ratings to incentivize quality (increasing app categories’ average user satisfaction rates by as much as 22%). This paper was accepted by Jayashankar Swaminathan, operations management.

Demystifying User’s Attachment of Smartphone Apps: A Value Orientation Perspective

This study aims to understand how features of an app can enhance smartphone app retention behavior among its users. Based on the literature review five features (relationship features, self-congruity, presentation style, novelty, and social connection) have been identified to influence the app self-connection, and app self-prominence, which in turn, eventually predict the affective commitment and behavioral attachment (i.e. word of mouth) of the app users. From a value orientation perspective, apps can be classified under three categories, namely, (i) Utilitarian apps, (ii) Hedonic apps, and (iii) Social network sites apps based on their usage and functionality. To enhance the generalizability of the study, data were collected from 750 active app users, i.e. 250 respondents from each category and a multi-group invariance analysis was performed to determine the significant effect of each category of apps on app retention behavior. The empirical results of the analysis highlighted the differential values exhibited by these five app features across three categories of smartphone apps. The findings escort the app-based marketers, and app developers to have a more sophisticated understanding of the respective category app features for desired business outcomes. The uniqueness of the present study is highlighted in its proposed integral framework that conceptualizes specific features of an app that the marketers can focus on to develop better-designed apps the users want.

Restrições socioambientais ao uso da terra e seu estado de antropização na UGRH do Paranapanema: subsídios à gestão territorial

Territorial planning and management are processes based on the organization of geographic space, which has been consolidated by means of common assumptions and rules that make living in society feasible. Through various legal mechanisms it is possible to determine areas with some kind of socio-environmental restrictions on land use. The present study proposes to spatialize in an integrated way different types of restrictive areas within the Paranapanema Water Resources Management Unit, namely Permanent Preservation Areas, Conservation Units, Priority Areas for Conservation of the Brazilian Biodiversity, Indigenous Lands, Natural-Cultural Heritage, Water Sources and Settlements. From the integration of these layers of information, through the algebra of maps, it was possible to determine up to six different classes of restrictions in the same unit area. The percentage of anthropization (especially areas agricultural activities areas) of each restrictive area mapped from land use data was also evaluated. Areas with up to six different classes of overlapping restrictions were identified, having been observed a high anthropization level throughout the study area, with approximately 80% of the territory by some agricultural use. This percentage was also observed in the category of APP of springs, which are important areas in terms of water resources. The areas with different overlapping and anthropized socio-environmental restrictions should be prioritized in environmental recovery actions, aiming, for example, the improvement of water quality in the Paranapanema UGRH.

You Are What Apps You Use: Demographic Prediction Based on User's Apps

Understanding the demographics of app users is crucial, for example, for app developers, who wish to target their advertisements more effectively. Our work addresses this need by studying the predictability of user demographics based on the list of a user's apps which is readily available to many app developers. We extend previous work on the problem on three frontiers: (1) We predict new demographics (age, race, and income) and analyze the most informative apps for four demographic attributes included in our analysis. The most predictable attribute is gender (82.3 % accuracy), whereas the hardest to predict is income (60.3 % accuracy). (2) We compare several dimensionality reduction methods for high-dimensional app data, finding out that an unsupervised method yields superior results compared to aggregating the apps at the app category level, but the best results are obtained simply by the raw list of apps. (3) We look into the effect of the training set size and the number of apps on the predictability and show that both of these factors have a large impact on the prediction accuracy. The predictability increases, or in other words, a user's privacy decreases, the more apps the user has used, but somewhat surprisingly, after 100 apps, the prediction accuracy starts to decrease.

Rating Prediction for Mobile Applications via Collective Matrix Factorization Considering App Categories

The rapid development of Internet and mobile communication, smart phone, tablet computer and other mobile devices become widespread, they have become an indispensable part of life, changing the human lifestyle, way of working and learning style, it brings great convenience and fun to us. This paper mainly discusses the status of personalized recommendation system, mobile application and analyse the main problems of recommendation system. This research proposes a rising mobile application domain, which uses web crawling technique to get explicit feedback from a real-world mobile application market. The explicit feedback is the users’ ratings data on Apps and the corresponding category information. To establish a mobile application recommendation model based on user interest and categorization information and propose a novel collective matrix factorization algorithm. Through the crawled data, the improved algorithm, and the traditional collaborative. The traditional mobile application recommendation systems are mostly based on keyword search, popularity, download usage and categories, thus they do not make personalized recommendation to users, and invalid to find applications that users are interested in. Through the user’s historical behaviour, a personalized recommendation system can be established to recommend the applications that fit user’s interests effectively. This paper is based on the collective matrix factorization using user interest such as user’s ratings and category information, designs a personalized recommendation system for mobile applications by using the relationship between them.

Managing the Versioning Decision over an App’s Lifetime

The authors investigate app publishers’ decisions to offer free, paid, or both versions of an app over an app’s lifetime by taking into account the interplays between the demand for the free and paid versions and publishers’ consideration of future profit streams. Their empirical analyses are based on a comprehensive model of publishers’ versioning decisions calibrated on a data set of 584 top-downloaded apps on Google Play. They find contemporaneous cannibalization but positive intertemporal cross-effects on new users’ demand between the two versions. In addition, the free version’s active user base and in-app purchase and advertising revenues are reduced by the presence of a paid version, but not vice versa. Among the three options, offering the paid version first is the most common optimal launch strategy and applies to 40% of apps in the data. The evolutionary patterns of optimal versioning decisions vary by app category and are related to apps’ abilities to monetize different revenue sources. This study provides insights on how to strategically manage the versioning decision over an app’s lifetime and shows how publishers can make their free version apps more profitable via the deployment or elimination of the paid version.

Balancing Public Health and Civil Liberties: Privacy Aspects of Contact-Tracing Technologies

Around the world, different jurisdictions, workplaces, and schools have taken wildly different approaches to contact tracing in response to COVID-19, striking varying balances between public health and safety and privacy and civil liberties. This article discusses contact-tracing technologies, with a particular focus on the privacy aspects of contact tracing and exposure-notification apps. It describes the three main categories of contact-tracing apps, namely centralized location-centralized Bluetooth-, and decentralized Bluetooth-based apps. This article observes lower adoption in the United States than in Europe and briefly considers what this microcosm could teach us about the intersection between privacy and technology.

Exposed! A case study on the vulnerability-proneness of Google Play Apps

Mobile applications are used for accomplishing everyday life activities, such as shopping, banking, and social communications. To leverage the features of mobile apps, users often need to share sensitive information. However, recent research demonstrated that most of such apps present critical security and privacy defects. In this context, we define as vulnerability-proneness the risk level(s) that users meet in downloading specific apps, to better understand whether (1) users select apps with lower risk levels and if (2) vulnerability-proneness of an app might affect its success. We use as proxy to measure such risk level the “number of different types of potential security issues exhibited by the app”. We conjecture that the vulnerability-proneness levels may vary based on (i) the types of data handled by the app, and (ii) the operations for which the app is supposed to be used. Hence, we investigate how the vulnerability-proneness of apps varies when observing (i) different app categories, and (ii) apps with different success levels. Finally, to increase the awareness of both users and developers on the vulnerability-proneness of apps, we evaluate the extent to which contextual information provided by the app market can be exploited to estimate the vulnerability-proneness levels of mobile apps. Results of our study show that apps in the Medical category exhibit the lowest levels of vulnerability-proneness. Besides, while no strong relations between vulnerability-proneness and average rating are observed, apps with a higher number of downloads tend to have higher vulnerability-proneness levels, but lower vulnerability-proneness density. Finally, we found that apps’ contextual information can be used to predict, in the early stages, the vulnerability-proneness levels of mobile apps.

Game Theory based Cyber-Insurance to Cover Potential Loss from Mobile Malware Exploitation

Potential for huge loss from malicious exploitation of software calls for development of principles of cyber-insurance. Estimating what to insure and for how much and what might be the premiums poses challenges because of the uncertainties, such as the timings of emergence and lethality of malicious apps, human propensity to favor ease by giving more privilege to downloaded apps over inconvenience of delay or functionality, the chance of infection determined by the lifestyle of the mobile device user, and the monetary value of the compromise of software, and so on. We provide a theoretical framework for cyber-insurance backed by game-theoretic formulation to calculate monetary value of risk and the insurance premiums associated with software compromise. By establishing the conditions for Nash equilibrium between strategies of an adversary and software we derive probabilities for risk, potential loss, gain to adversary from app categories, such as lifestyles, entertainment, education, and so on, and their prevalence ratios. Using simulations over a range of possibilities, and using publicly available malware statistics, we provide insights about the strategies that can be taken by the software and the adversary. We show the application of our framework on the most recent mobile malware data (2018 ISTR report—data for the year 2017) that consists of the top five Android malware apps: Malapp, Fakeinst, Premiumtext, Maldownloader , and Simplelocker and the resulting leaked phone number, location information, and installed app information. Uniqueness of our work stems from developing mathematical framework and providing insights of estimating cyber-insurance parameters through game-theoretic choice of strategies and by showing its efficacy on a recent real malicious app data . These insights will be of tremendous help to researchers and practitioners in the security community.

Medical Apps and the Gray Zone in the COVID-19 Era: Between Evidence and New Needs for Cybersecurity Expansion.

The study focuses on emerging problems caused by the spread of medical apps. Firstly, it reviews the current role of cybersecurity and identifies the potential need to widen the boundaries of cybersecurity in relation to these apps. Secondly, it focuses on the pivotal device behind the development of mHealth: the smartphone, and highlights its role and current potential for hosting wearable medical technology. Thirdly, it addresses emerging issues regarding these apps, which are in a gray zone. This is done through an analysis of the important positions of scholars, and by means of a survey report on the increased use of various categories of apps during the COVID-19 pandemic, highlighting an accentuation of the problem. The study ends by explaining the reflections and proposals that emerged after performing the analysis.

Studying Ad Library Integration Strategies of Top Free-to-Download Apps

In-app advertisements have become a major revenue source for app developers in the mobile app ecosystem. Ad libraries play an integral part in this ecosystem as app developers integrate these libraries into their apps to display ads. In this paper, we study ad library integration practices by analyzing 35,459 updates of 1,837 top free-to-download apps of the Google Play Store. We observe that ad libraries (e.g., Google AdMob) are not always used for serving ads -- 22.5% of the apps that integrate Google AdMob do not display ads. They instead depend on Google AdMob for analytical purposes. Among the apps that display ads, we observe that 57.9% of them integrate multiple ad libraries. We observe that such integration of multiple ad libraries occurs commonly in apps with a large number of downloads and ones in app categories with a high proportion of ad-displaying apps. We manually analyze a sample of apps and derive a set of rules to automatically identify four common strategies for integrating multiple ad libraries. Our analysis of the apps across the identified strategies shows that app developers prefer to manage their own integrations instead of using off-the-shelf features of ad libraries for integrating multiple ad libraries. Our findings are valuable for ad library developers who wish to learn first hand about the challenges of integrating ad libraries.

Analysis and Classification of Mobile Apps Using Topic Modeling: A Case Study on Google Play Arabic Apps

Mobile app stores provide an extremely rich source of information on app descriptions, characteristics, and usage, and analyzing these data provides insights and a deeper understanding of the nature of apps. However, manual analysis of this vast amount of information on mobile apps is not a simple and straightforward task; it is costly in terms of human effort and time. Computational methods such as topic modeling can provide an efficient and satisfactory approach to mobile app information analysis. Topic modeling is a type of statistical modeling technique for discovering abstract topics that occur in a set of documents. This study explores the relationship between features of Arabic apps and investigates how well the current predefined Google Play app categories represent the type and genre of Arabic mobile apps. Based on the textual app description analysis, we aim to design and develop a sustainable classification system using the Latent Dirichlet Allocation (LDA) method of topic modeling in order to cover the Arabic apps classification in Google Play app store. Our study supports the hypothesis that the textual app descriptions are effective in suggesting new categories for Arabic mobile apps in Google Play app store. Also, the results indicated that the current classification on Google Play app store is not suitable for our case study “Arabic apps,” as well as it is not sustainable, as it can not cover the new app types including Arabic apps. This study offers an important contribution to Arabic app analysis and design, to improve app search and exploration in several domains such as business, marketing, and technical development. Furthermore, it provides insights for the future of Arabic app research and provides guidance for the development of an Arabic app dashboard that will support users on how to select an app based on their specific needs.

Switching decision, timing, and app performance: An empirical analysis of mobile app developers’ switching behavior between monetization strategies

Mobile application developers switch their monetization strategies for strategic purposes. We investigate the determinants and consequences of app developers’ switching behavior between paid and free strategies over time. Using large-scale daily app rank data from the iOS App Store, our estimations reveal that the prior performance of an app and its duration of staying in a monetization strategy significantly impact the developers’ decision of switching. We also find that better-performing apps tend to have longer stays in the paid strategy, whereas developers tend to switch their monetization strategy to free when the performance of a paid app declines. Furthermore, app developers’ switching behavior and duration of the app in each strategy have significant effects on the subsequent app performance in the App Store. Finally, we find the impact of switching behavior varies by app category. Managerial implications with estimated economic outcomes on strategic moves by app developers towards higher app performance are provided.

HybriDroid: an empirical analysis on effective malware detection model developed using ensemble methods

Malware detection from the smartphone has become a challenging issue for academicians and researchers. In this research paper, we applied five distinct machine learning algorithms and three different ensemble methods to develop a model for detecting malware from an Android-based smartphone. In this study, we proposed a framework that helps in selecting the right sets of the feature with an aim to improve the performance of the malware detection models. The proposed malware detection framework is then validated by considering two distinct performance parameters, i.e., accuracy and F-measure as a benchmark to detect malware from real-world apps. We performed an empirical study on thirty different categories of Android apps. The experimental data set consists of 1,94,659 benign apps and 67,538 malware apps that are collected from different promised repositories. Empirical results reveal that the models developed by using the proposed feature selection framework are able to detect more malware-infected apps when compared to all extracted feature sets. Moreover, the malware detection model build by using nonlinear ensemble decision tree forest (NDTF) approach is achieved a detection rate of 98.8%. In addition to that, the proposed malware detection framework is more effective in detecting malware-infected apps as compared to different anti-virus scanners and different frameworks or approaches developed in the literature.

The Adoption of Mobile Payment Technologies, Social Interactive Consumer-Oriented Applications, and Online Purchasers’ Decision-Making Process

Research background: This paper analyzes the outcomes of an exploratory review of the current research on the relationship between the global adoption of mobile payment technologies, social interactive consumer-oriented applications, and online purchasers’ decision-making process. Purpose of the article: The data used for this study was obtained and replicated from previous research conducted by Econsultancy and Statista. We performed analyses and made estimates regarding mobile e-commerce sales worldwide, frequency of mobile retail app usage according to U.S. smartphone shoppers, how well organizations understand the customer journey for certain audiences, share of Internet users who are likely to use mobile payments on their smartphone in the next year (by country), and time spent per mobile app category. Methods: Data collected from 6,200 respondents are tested against the research model by using structural equation modeling. Findings & Value added: The advent of smartphones redesigns the routine of shopping, thus altering the agency of users. Retailers have instant access to data on the geographical position of users that can be employed in addition to other information to regulate the decision-making process. Perceived effortlessness in utilization of the smartphone is somewhat similar for various mobile shopping application settings. The degree to which mobile purchasing applications are time critical and location sensitive may differ substantially. Mobile retailers can make public the personal, collaborative, and instantaneous buying experience that mobile shopping can offer to customers.

Availability of Spanish-Language Medical Apps in Google Play and the App Store: Retrospective Descriptive Analysis Using Google Tools

BackgroundThe number of medical and health apps in the App Store and Google Play repositories has been increasing in the recent years, and most of these apps are in English. However, little is known about the domain of Spanish health apps and their evolution.ObjectiveThe aim of this study was to perform a retrospective descriptive analysis of medical apps for patients in the Spanish language by using Google search tools over a 5-year period and to compare the results by using a reproducible methodology to obtain a better knowledge of the medical apps available in the Spanish Language.MethodsOver a 5-year period, medical apps were catalogued using a Google-based methodology. Keywords of the first 14 categories of the International Classification of Diseases, Tenth Revision, were selected, and in December of each year, searches of the URLs of Google Play and the App Store were conducted using Google Advanced Search. The first 10 results were taken, and apps meeting the inclusion criteria were selected and rated with the iSYScore method.ResultsOut of a sample of 1358 apps, 136 met the inclusion criteria. The 3 main categories of the medical apps were in the fields of endocrinology (diabetes), respiratory (chronic obstructive pulmonary disease, asthma, and allergies), and neurology (multiple sclerosis, Parkinson disease, and Alzheimer disease). Few apps were maintained over the 5 years. Only 10 of the 136 apps were maintained for 3 years or more. There was a large number of original apps in other languages that were translated into Spanish (56/136, 41.2%). In the last year of the study, the main reason (73/280, 26.1%) for discarding an app was the date of the last update.ConclusionsThe market of Spanish apps is poor; only few apps have appeared repeatedly over 5 years. Differences were found with the international market in terms of apps related to mental health, heart and circulatory system, and cancer, and coincidences were found in the relevance of apps for diabetes control.

The Effects of COVID-19 on Mobile App Usage

How have social distancing restrictions influenced customers' uses of mobile applications ("apps") during the Covid-19 pandemic? In this research, we examine a proprietary data set that contains usage information from around 280 mobile apps between January 2019 and July 2020 in 30 major Chinese cities. We find that the effect of social distancing policy on the number of active users depends on the nature of the app. Using regression discontinuity in time (RDiT), we find that social distancing policy increased the number of active users by an average of 21% across all apps. We stratify the data by two primary app categories: Life Service apps (which involve offline human-human or human-product contact), and Virtual Service apps (which involve no offline contact), and find that within the Life Service category, social distancing policy helped popular/independent apps the most (72% increase in the number of active users), followed by niche/independent apps, while hurt popular/platform apps and niche/platform apps. For apps in the Virtual Service category, niche/independent apps and popular/platform apps benefited the most from social distancing policy, ranging roughly from 12% to 54%. Furthermore, we find that in cities with higher density and more migrants (vs. cities with lower density and fewer migrants), the number of active users increased by 37% and 34% (vs. 8% and 11%) respectively. The results support two underlying mechanisms of social distancing policy: social deprivation (i.e., the feeling of being deprived of social interactions) and offline risk (i.e., the risk of exposure to the coronavirus).