App Ranking Research Articles

E-SERS: An Enhanced Approach to Trust-Based Ranking of Apps

The number of mobile applications (“Apps”) has grown significantly in recent years. App Stores rank/recommend Apps based on factors such as average star ratings and the number of installs. Such rankings do not focus on the internal artifacts of Apps (e.g., security vulnerabilities). If internal artifacts are ignored, users may fail to estimate the potential risks associated with installing Apps. In this research, we present a framework called E-SERS (Enhanced Security-related and Evidence-based Ranking Scheme) for comparing Android Apps that offer similar functionalities. E-SERS uses internal and external artifacts of Apps in the ranking process. E-SERS is a significant enhancement of our past evidence-based ranking framework called SERS. We have evaluated E-SERS on publicly accessible Apps from the Google Play Store and compared our rankings with prevalent ranking techniques. Our experiments demonstrate that E-SERS, leveraging its holistic approach, excels in identifying malicious Apps and consistently outperforms existing alternatives in ranking accuracy. By emphasizing comprehensive assessment, E-SERS empowers users, particularly those less experienced with technology, to make informed decisions and avoid potentially harmful Apps. This contribution addresses a critical gap in current App-ranking methodologies, enhancing the safety and security of today’s technologically dependent society.

Measuring and Understanding Crowdturfing in the App Store

Application marketplaces collect ratings and reviews from users to provide references for other consumers. Many crowdturfing activities abuse user reviews to manipulate the reputation of an app and mislead other consumers. To understand and improve the ecosystem of reviews in the app market, we investigate the existence of crowdturfing based on the App Store. This paper reports a measurement study of crowdturfing and its reviews in the App Store. We use a sliding window to obtain the relationship graph between users and the community detection method to binary classify the detected communities. Then, we measure and analyze the crowdturfing obtained from the classification and compare them with genuine users. We analyze several features of crowdturfing, such as ratings, sentiment scores, text similarity, and common words. We also investigate which apps crowdturfing often appears in and reveal their role in app ranking. These insights are used as features in machine learning models, and the results show that they can effectively train classifiers and detect crowdturfing reviews with an accuracy of up to 98.13%. This study reveals malicious crowdfunding practices in the App Store and helps to strengthen the review security of app marketplaces.

The Role of M-Government Application in the Saudi Health Sector in Light of the COVID-19 Pandemic: A Review

The Kingdom of Saudi Arabia’s government had taken an interest in digital health. This helped to speed up the development of services needed to confront the Pandemic. This paper aims to review the role of free mobile health apps available on Google Play and the App Store, and used in Saudi Arabia during the COVID-19 outbreak. This study relied on a review of the mobile applications currently used for COVID-19. Six applications were selected that included the specific search criteria. The study found that apps have met the inclusion criteria. The following characteristic of the applications have been described: Application overview (user rating, platform (Android, IOS), language, app ranking in the store). Also, application analysis and service criteria shown based on the functions that match the specified criteria (social distancing, isolation/quarantine, awareness). We noticed a redundancy of services between 6 applications. In conclusion, it proposes the design and integration of services through a mobile health application with most of the features and functions analysed in this study.

모바일 앱 포트폴리오 전략이 모바일 개발자의 성과에 미치는 영향

What mobile app development strategy should the developer consider to survive and grow in hypercompetitive mobile application market? Mobile app market is one of the most rapidly growing segment in software market. Although mobile app market is a promising area, this positive perspective is not consistently applicable to app developers. In extremely competitive mobile app market, application developers would need to carefully consider and formulate app development strategies for success. To define the strategies, product portfolio management composing with app development speed, diversification speed, and switch speed is introduced. Quantitative analysis validating the statistical relationship between portfolio strategies and app developer’s financial performance is conducted. For this analysis, data from app developers with worldwide cumulative revenue estimate of over $100,000 were collected. As a result of data analysis, positive relationship between app development speed and developer’s financial performance was supported whereas diversification speed and switch speed had negative relationship with revenue. Theoretically, this research differs from the previous mobile app related studies. While previous studies used app ranking or user ratings as proxy for app performance, this study uses estimated sales volume as dependent variable which reflects more accurate and empirical performance.

Ranking of Financial Apps Using Fuzzy Ahp and Fuzzy Marcos: An Application of Multi-Criteria Decision-Making (Mcdm) Techniques

Ranking of Financial Apps Using Fuzzy Ahp and Fuzzy Marcos: An Application of Multi-Criteria Decision-Making (Mcdm) Techniques

Make the apps stand out: discoverability and perceived value are vital for adoption

PurposeFrom an environmental psychology perspective, we aim to uncover the role that app discoverability facilitators play in enabling the various perceived values (e.g. social, information and hedonic) necessary for app adoption.Design/methodology/approachA survey study was conducted and data was analyzed using partial least squares structural equation modeling (PLS-SEM).FindingsResults show that certain consumer review elements (i.e. review quality, review quantity and app ranking), peer influence and app developers' reputation — to varying degrees — influence the three perceived values, which subsequently affect users' app adoption intention. The three perceived values mediate the relationship between app discoverability facilitators and users' app adoption intention.Practical implicationsApp store managers and developers should make a greater effort to effectively optimize discoverability and product differentiation.Originality/valueGuided by environmental psychology, we confirm the importance of app discoverability facilitators regarding their influence on users' general perceptions of an app (e.g. the three perceived values). We also uncover the differentiated effect of the three perceived values on app adoption intention.

Android App Rating Classification on Google Play Store Using Random Forest Algorithm with SQL Server Preprocessing

The increasing number of Android applications available on the Google Play Store with the benefits the developers get has attracted the attention of many Android application developers. To benefit from developing Android apps, one way is to know the characteristics of highly rated apps on the Google Play Store. This research will investigate the features of size, installs, reviews, type (free / paid), rating, category, content rating, and price on applications on the Google Play Store to determine the characteristics of high-rated applications. This study uses the Random Forest algorithm to identify the most influential features in high ranking applications on the Google Play Store. At the preprocessing stage, this research uses data cleaning methods and data reduction using SQL Server. This study uses feature important to find out the attributes that most influence the high ranking of Android apps on the Google Play Store. To classify high-ranking applications, the authors use 8-fold cross validation using the Random Forest algorithm and get better results than the Gradient Boost, K-NN, and Decision Tree algorithms with an accuracy of 83%. The results of the Random Forest algorithm also have better performance than the algorithm from the previous research conclusions, with a 0.8% increase in accuracy. To classify high-ranking applications, the authors use 8-fold cross validation using the Random Forest algorithm and get better results than the Gradient Boost, K-NN, and Decision Tree algorithms with an accuracy of 83%. The results of the Random Forest algorithm also have better performance than the algorithm from the previous research conclusions, with a 0.8% increase in accuracy. To classify high-ranking applications, the authors use 8-fold cross validation using the Random Forest algorithm and get better results than the Gradient Boost, K-NN, and Decision Tree algorithms with an accuracy of 83%. The results of the Random Forest algorithm also have better performance than the algorithm from the previous research conclusions, with a 0.8% increase in accuracy.

Analysis Of Physical Activity-Related Smartphone Apps For Behavior Change Potential

The majority of Americans do not meet physical activity (PA) guidelines. Adoption and adherence may be improved with personalized intervention. Smartphone app use has increased exponentially in the past decade, extending to many areas of life and crossing socioeconomic status, ethnicity, and gender. Free smartphone apps that address habitual PA may be a useful tool, but their potential to support PA behavior change is not well understood. PURPOSE: To evaluate top-ranked free health and fitness apps for PA behavior change potential and to determine the relationship between this evaluation and consumer perceptions of quality, such as app store ranking and consumer ratings. METHODS: The top 100 "free" health and fitness apps in the US iOS app store were screened for rinclusion. Apps were excluded if the descriptions did not address any aspects of PA or fitness; required an external device, purchase/subscription after a trial period, or a health club membership or specific health insurance. Apps were also excluded if their descriptions was not written in English. Two investigators downloaded and evaluated the remaining apps that met the inclusion criteria using the App Behavior Change Scale (ABACUS). Apps were scored on 21 strategies within 4 categories (knowledge/information, goals/planning, feedback/monitoring, and actions), and prevalence in the sample was calculated. Pearson correlations were estimated for the relationship between ABACUS total scores and relative app ranking, as well as consumer rating. Significance was set at p < 0.05. RESULTS: 23 apps met the inclusion criteria. 100% offered some form of personalization; 96% incorporated self-monitoring and/or rehearsal; and 70% included options for goal-setting. Only 13% recommended restructuring of the environment, and 4% provided advice on distraction or avoidance. There were no significant correlations between ABACUS score and store ranking (r2= -0.15, p = 0.49) or consumer rating (r2= -0.26, p = 0.91) . CONCLUSIONS: Free apps may have the potential to support PA behavior change, especially in the areas of planning and self-monitoring. There is no correlation between the total number of evidence-based behavior change strategies and app popularity, however, so it may be beneficial for wellness professionals to counsel patients and clients on app choice.

Herding and Software Adoption: A Re-Examination Based on Post-Adoption Software Discontinuance

ABSTRACT Informational cascades are theorized as an underlying mechanism of herding. That is, an individual, having observed the actions of those ahead of him/her, chooses to follow the behavior of the preceding individuals even though his/her private information suggests other options. Empirical identification of informational cascades is challenging because individual users’ private information is unobservable. Our study utilizes a unique data set on post-adoption discontinuance of app usage to revisit herding and informational cascades in software adoption. We find that with the download of apps being controlled, a higher software ranking is associated with more post-adoption discontinuance of app usage, which empirically illustrates the decision deficiency of following others’ observed behavior in adopting popular software apps and supports the theoretical perspective of informational cascades. We further show that the association between app ranking and post-adoption discontinuance is stronger for apps with higher ratings and with higher complexity levels. Moreover, as apps become more complex in the app life cycle, updated app versions with a higher level of complexity are associated with a weaker relationship between app ranking and post-adoption discontinuance. Our study contributes to the literature by confirming the informational cascades effect and its interaction with other informational mechanisms (e.g., user rating) and software internal feature (e.g., product complexity) in software adoption. The findings help software vendors gain insights in users’ herding behavior in software adoption and optimize their software releasing strategies and promotional effort allocation.

Digital Tools Fill a Gap in Mental Health Screening and Support, Particularly for Women Lacking Strong Social Networks

Background Roughly 11% of women suffer from postpartum depression nationwide; however, many believe the condition to be widely underreported, in part due to inadequate screening and stigma associated with the condition. Social support networks can help to prevent or mitigate symptoms related to postpartum depression. Single mothers tend to suffer from this condition at a higher rate than married women as they tend to have weaker social networks compared to married women. Objective The primary objective ws to determine whether gaps exist in mental health screening and whether digital screening tools can help to fill these gaps. The secondary objective ws to determine whether digitally delivered support proves to be more or less beneficial to subsets of women, namely based on their marital status. Methods A survey about mental health history, support, experience with mental health screeners, and characteristics of social networks was sent by email to users of the Ovia Fertility, Ovia Pregnancy, and Ovia Parenting mobile apps. Respondents were all 18 years of age or older and living in the United States. The study was granted exemption by our institutional review board. Results Of the 2016 respondents, 39% reported that they were never screened by their healthcare provider for mental health conditions (26% of women with children and 52% of women without children). Among women who reported never being screened by a healthcare provider, 17% reported that they have completed at least one of the screeners (PHQ-9 or Edinburgh Postnatal Depression Scale [EPDS]) in an Ovia mobile app. Of the 2016 respondents, 86% reported being married or in a domestic partnership. Among the single respondents, 32% reported either having children, being pregnant, or currently trying to conceive. More single women who have children, are pregnant, or are actively trying to conceive reported that they would feel most supported by a mobile appl (namely, one of Ovia Health’s three mobile apps) and to seek treatment for mental health concerns compared to married women (19% compared to 14% of married women; P=.03). Additionally, single women who have children, are pregnant, or are actively trying to conceive reported more often than married women that they feel their mental health is best supported by a mobile appl (16% compared to 10% of married women; P=.007). However, both groups of women selected their healthcare provider and their friends/family as the first and second ranking support systems for both seeking mental health treatment and for mental health related support, with the mobile app ranking last. Conclusions Screening for mental health conditions during the reproductive health journey is lacking. Digital solutions that deliver clinically validated screening tools help to screen women who are missed in a clinical setting. Women who report being single throughout parenting, pregnancy, or while trying to conceive find more value in mobile app–provided mental health support compared to married women. These findings highlight two gaps that digital technologies, like Ovia Health, can fill: low mental health screening rates during reproductive years and suboptimal social systems.

Detecting Promotion Attacks in the App Market Using Neural Networks

App markets play an important role in distributing various apps to mobile users. The app market vendors provide reputation systems to assist users in finding useful and reputable apps by ranking them. Unfortunately, there are signs that an app's ranking can be easily manipulated, which causes unfair competition for those highly ranked ones. Here we propose a novel approach based on deep learning to detect such malicious ranking manipulations. The proposed neural network has a novel architecture that is able to incorporate a variety of features designed from the publicly available application information in the app market. We have conducted extensive experiments as well as individual case analysis and the results demonstrate the effectiveness of our proposed approach.

Uncovering top-ranking factors for mobile apps through a multimethod approach

The increasing computational power of mobile devices and the advancements in network communications are enabling the emergence of new mobile services. Developers have created many mobile applications (mobile apps) to fulfill a wide range of personal and professional user needs. The present study aims to answer the following research question: what are the factors that influence an app's ranking and success? To answer this question, we define a set of antecedents that may explain the top rank of an app. We use a sample of 500 of Apple's top grossing apps to analyze the top 50 and bottom 50 apps. We then use a multivariate logistic regression to examine if factors such as user rating, category popularity, diversity as measured by the number of languages supported, package size, and release date are determinants of an app's success. We also apply a fuzzy-set qualitative comparative analysis (fsQCA) to find the existence of more causal paths for the mobile app's success. Multivariate results indicate that category popularity, diversity (number of languages supported), package size, and app release date are all factors that increase the probability that an app will be ranked inside the top 50. Nevertheless, contrary to our prediction, a high user rating is negatively associated with an app's success. The results of the fsQCA show that the importance of an app's attributes, functionalities, and longevity surpasses the importance of the user rating in explaining the app's success.

Herding and Software Adoption: A Reexamination Based on Post-Adoption Software Discontinuance

Informational cascades are theorized as an underlying mechanism of herding. That is, an individual, having observed the actions of those ahead of him/her, chooses to follow the behavior of the preceding individuals even though his/her private information suggests other options. Empirical identification of informational cascades is challenging because individual users’ private information is unobservable. Our study utilizes a unique data set on post-adoption discontinuance of app usage to revisit herding and informational cascades in software adoption. We find that with the download of apps being controlled, a higher software ranking is associated with more post-adoption discontinuance of app usage, which empirically illustrates the decision deficiency of following others’ observed behavior in adopting popular software apps and supports the theoretical perspective of informational cascades. We further show that the association between app ranking and post-adoption discontinuance is stronger for apps with higher ratings and with higher complexity levels. Moreover, as apps become more complex in the app life cycle, updated app versions with a higher level of complexity are associated with a weaker relationship between app ranking and post-adoption discontinuance. Our study contributes to the literature by confirming the informational cascades effect and its interaction with other informational mechanisms (e.g., user rating) and software internal feature (e.g., product complexity) in software adoption. The findings help software vendors gain insights in users’ herding behavior in software adoption and optimize their software releasing strategies and promotional effort allocation.

A data reduction scheme for active authentication of legitimate smartphone owner using informative apps ranking

Smartphones are vulnerable to fraudulent use despite having strong authentication mechanisms. Active authentication based on behavioral biometrics is a solution to protect the privacy of data in smart devices. Machine-learning-based frameworks are effective for active authentication. However, the success of any machine-learning-based techniques depends highly on the relevancy of the data in hand for training. In addition, the training time should be very efficient. Keeping in view both issues, we’ve explored a novel fraudulent user detection method based solely on the app usage patterns of legitimate users. We hypothesized that every user has a unique pattern hidden in his/her usage of apps. Motivated by this observation, we’ve designed a way to obtain training data, which can be used by any machine learning model for effective authentication. To achieve better accuracy with reduced training time, we removed data instances related to any specific user from the training samples which did not contain any apps from the user-specific priority list. An information theoretic app ranking scheme was used to prepare a user-targeted apps priority list. Predictability of each instance related to a candidate app was calculated by using a knockout approach. Finally, a weighted rank was calculated for each app specific to every user. Instances with low ranked apps were removed to derive the reduced training set. Two datasets as well as seven classifiers for experimentation revealed that our reduced training data significantly lowered the prediction error rates in the context of classifying the legitimate user of a smartphone.

Free Diabetes Smartphone Apps—Downloads and Rankings 2016 vs. 2017

Purpose: Do the frequency of downloads of free diabetes smartphone droid apps change from year to year? Methods: Using Google playstore, we found the most popular apps in 2017. For 2016 apps, we used our 2016 data 1 from the Google playstore. The 11 most popular free apps in 2016 and 2017 were compared. Results: The change in app ranking were associated with "pay for premium features (PPF)" in 2017. The app, mySugr remains no. 1 in terms of number of downloads, due to fewest changes in PPF. Diabetes M went from no. 5 in 2016 to no. 8 in 2017, due to PPF. Glucose Buddy remained at no. 4 in the two years. Blood Glucose Tracker went from 2 to 11 in two years due to high cost PPF. Of the most popular apps in 2017, Diabetes Plus is only free on android. In 2016, only 2 are for children:Carb Counting with Lenny and mySugr Junior. As of 2017, mySugr Junior no longer exists. None of the 2017 or 2016 apps have glycemic index. All apps in 2017 require email login. 4/11 apps record height and weight in 2017. 3/11 apps do not record blood pressure in 2017. 5/6 apps were journal based vs. 12/12 apps in 2017. Conclusion: The popularity of apps change greatly in 24 months. Patients and MDs have a common goal in seeking useful mobile apps for diabetes self-management. Disclosure G. Wu: None. V. Ngo: None.

Do Free Smartphone Apps Have Features of the Diabetes Self-Management Education/Support Guidelines?

Purpose: Assessing free diabetes smartphone apps in 2016 and 2017 for popularity and features of the Diabetes Self Management/Education Guidelines (DSME) from the ADA. Methods: Google analytics were used to evaluate the 11 most popular apps for 2016 and 2017. The apps were then assessed with the ADA's DSME guidelines. The 2016 app ranking was taken from our unpublished data. 1 Results: While 8/11 had HbA1c in 2016, this category improves to 100% in 2017. In 2016 v 2017: 6/11 vs. 10/11 used calorie counting. For triglycerides and lipids, both in 2016 and 2017, only one app features this category. For psychological moods, there were two apps with this feature in 2016 but only one in 2017. Goal setting in 2016 v 2017: 4 vs. 5. Conclusions: The apps from 2017 are greatly improved over 2016. All except one app show cal/carbs, HbA1c/FBS, medical diary/log. However there is still room for improvement for the categories of Triglycerides/lipids, Psychological mood, BMI and Goal setting. We, as physicians, look forward to more robust free apps for diabetic patients. Reference: Gao M, Wu G, Bautista BM et al. "2016 Diabetes Smartphone Apps and Diabetes Self Management Education/Support Guidelines. Endocrine Society poster presentation, Orlando FL, April 3, 2017. Disclosure G. Wu: None. V. Ngo: None.

Detecting Malicious Facebook Applications

With 20 million installs a day [1], third-party apps are a major reason for the popularity and addictiveness of Facebook. Unfortunately, hackers have realized the potential of using apps for spreading malware and spam. The problem is already significant, as we find that at least 13% of apps in our dataset are malicious. So far, the research community has focused on detecting malicious posts and campaigns. In this paper, we ask the question: given a Facebook application, can we determine if it is malicious? Our key contribution is in developing FRAppE—Facebook‟s Rigorous Application Evaluator— arguably the first tool focused on detecting malicious apps on Facebook. To develop FRAppE, we use information gathered by observing the posting behavior of 111K Facebook apps seen across 2.2 million users on Facebook. First, we identify a set of features that help us distinguish malicious apps from benign ones. For example, we find that malicious apps often share names with other apps, and they typically request fewer permissions than benign apps. Second, leveraging these distinguishing features, we show that FRAppE can detect malicious apps with 99.5% accuracy, with no false positives and a low false negative rate (4.1%). Finally, we explore the ecosystem of malicious Facebook apps and identify mechanisms that these apps use to propagate. Interestingly, we find that many apps collude and support each other; in our dataset, we find 1,584 apps enabling the viral propagation of 3,723 other apps through their posts. Long-term, we see FRAppE as a step towards creating an independent watchdog for app assessment and ranking, so as to warn Facebook users before installing apps.

CRSPR: PageRank for Android Apps

With the sharp increase in mobile apps, modular design and functional reuse are commonly adopted. The inter-component communication (ICC) mechanism in Android allows apps to exchange data with other apps and components, resulting in large amounts of security issues, such as component hijacking vulnerabilities, privilege escalation and spoofing attacks. Although ICC has been extensively studied in previous work, none of the previous approaches is practically scalable to simultaneously analyze a large number of Android apps, giving the combinational explosion of possible inter-component (and inter-app) communications. In this paper, we first propose an explorative study to analyze the ICC-based interaction for a large amount of Android apps. Then we propose CRSPR, a PageRank-like topic-aware app ranking approach to highlight influential Android apps for ICC analysis. The experimental results show that CRSPR is better than the basic counting approach as well as the traditional PageRank-based approach, which further demonstrate that CRSPR is useful for highlighting influential Android apps.

Fraud Detection of Mobile App Ranking

Fraud Detection of Mobile App Ranking

모바일 애플리케이션 다운로드에 영향을 미치는 요인에 관한 연구

스마트폰 시장이 성장하면서 모바일 애플리케이션 시장은 잠재력 있는 새로운 시장으로 평가받고 있다. 애플스토어 같은 애플리케이션 시장에서 많은 구매자들이 모바일 애플리케이션을 다운로드 받고 있지만, 아직 소비자의 애플리케이션 다운로드에 대한 주요 연구는 없는 실정이다. 그러므로 이 연구는 스마트폰용 모바일 애플리케이션 다운로드에 영향을 미치는 요소에 대해서 연구를 하자고 한다. 앱 순위는 앱 다운로드에 유의한 영향을 미치는 것으로 나타났다. 가격은 유의적인 영향을 발견할 수 없었다. 앱 유형, 개발자 경험, 지역성은 애플 앱스토어에서만 유료앱 다운로드에 유의적인 영향을 미치는 것으로 나타났다. 이 연구가 모바일 애플리케이션 비즈니스 구축 및 모바일 애플리케이션 시장 활성화 전략 기여를 할 수 있을 것으로 기대된다. Mobile applications are significantly impacting people's smart phone using behavior and mobile industry value chain. By examining sources of data on mobile applications use, this study presents evidence about what factors would affect the amount of apps download, which may be useful to guide app designers and publishers to develop more persuasive new apps and marketing strategies. The results indicated app ranking had effect on download amount of apps in both Android app market and Apple app store, while prices of apps had no impact on the amount of download. App type, developer experience, and locality had effect on the amount of download only for paid apps in Apple app store.