A network traffic-aware mobile application recommendation system based on network traffic cost consideration

A large amount and different types of mobile applications (or apps) are being offered to end users via app markets. These apps normally generate network traffic, which will consumes users' mobile data plan and may even cause potential security issues. However, the amount and type of network traffic generated by a mobile app in the wild is still poorly understood due to the lack of a systematic measurement methodology. In this paper, we first measure and analyze network traffic cost of Android apps in the official Android markets. Based on the results, we find that the apps from different categories have different traffic costs. In particular, there is a remarkable difference among the apps with similar functionality in terms of network traffic cost. Then, we add metrics of traffic cost into our app recommendation algorithm, which differs from the conventional app recommendation approaches. Experimental results show that the proposed recommendation algorithm can effectively help mobile app users avoid various potential security and privacy risks brought by the unnecessary network traffic consumption.

Among several traditional and novel mobile app recommender techniques that utilize a diverse set of app-related features (such as an app’s Twitter followers, various version instances, etc.), which app-related features are the most important indicators for app recommendation? In this paper, we develop a hybrid app recommender framework that integrates a variety of app-related features and recommendation techniques, and then identify the most important indicators for the app recommendation task. Our results reveal an interesting correlation with data from third-party app analytics companies; and suggest that, in the context of mobile app recommendation, more focus could be placed in user and trend analysis via social networks.

Purpose Explore smartphone apps that may be recommended by clinicians for the self-management of low back pain. Methods Prospectively registered systematic assessment of self-contained apps for self-management of low back pain on the Google Play and Apple App stores (Oceania), including ≥1 NICE low back pain and sciatica clinical guideline-recommended component and functioning without health professional input. Outcomes were quality (Mobile App Rating Scale; MARS), and self-management (Self-Management Support Checklist; SMS-14) and behaviour change potential (App Behaviour Change Scale; ABACUS). Results 25 apps were included. The average quality of included apps was acceptable (Mean MARS score of 3.9 out of a maximum possible 5). The self-management support and behaviour change potential of included apps appeared low (mean SMS-14 score was 3.4/14; mean ABACUS score was 5.4/21). The apps showed no significant correlation between app consumer ratings and MARS scores. App quality was significantly correlated with app price (p = 0.049) but not consumer ratings, however, these findings were based on a small number of studies and the overall model was not significant. Conclusions Smartphone apps for the self-management of low back pain are of average to good quality, with questionable potential for self-management and behaviour change. Clinicians should consider that few apps were designed to specifically incorporate self-management support and behaviour change potential when recommending apps to clients. Further development in these areas of app design would be of benefit. Implications for Rehabilitation Smartphone apps have the potential to improve LBP self-management outcomes, however, apps are not well regulated and the quality of information and advice provided is often low quality. The findings from this systematic assessment indicate that LBP self-management apps have limited self-management support and behaviour change potential features. Recommendation of current smartphone apps for LBP should take into consideration that although apps are of acceptable quality they are not specifically designed with self-management support and behaviour change principles. App quality does not appear to be associated with consumer ratings, however, there may be a positive association between app price and quality which requires further investigation.

Popularity of mobile apps is traditionally measured by metrics such as the number of downloads, installations, or user ratings. A problem with these measures is that they reflect usage only indirectly. Indeed, retention rates, i.e., the number of days users continue to interact with an installed app, have been suggested to predict successful app lifecycles. We conduct the first independent and large-scale study of retention rates and usage trends on a dataset of app-usage data from a community of 339,842 users and more than 213,667 apps. Our analysis shows that, on average, applications lose 65% of their users in the first week, while very popular applications (top 100) lose only 35%. It also reveals, however, that many applications have more complex usage behaviour patterns due to seasonality, marketing, or other factors. To capture such effects, we develop a novel app-usage trend measure which provides instantaneous information about the popularity of an application. Analysis of our data using this trend filter shows that roughly 40% of all apps never gain more than a handful of users ( Marginal apps). Less than 0.1% of the remaining 60% are constantly popular ( Dominant apps), 1% have a quick drain of usage after an initial steep rise ( Expired apps), and 6% continuously rise in popularity ( Hot apps). From these, we can distinguish, for instance, trendsetters from copycat apps. We conclude by demonstrating that usage behaviour trend information can be used to develop better mobile app recommendations.

This paper focuses on the evaluation and recommendation of healthcare applications in the mHealth field. The increase in the use of health applications, supported by an expanding mHealth market, highlights the importance of this research. In this study, a data set including app descriptions, ratings, reviews, and other relevant attributes from various health app platforms was selected. The main goal was to design a recommendation system that leverages app attributes, especially descriptions, to provide users with relevant contextual suggestions. A comprehensive pre-processing regime was carried out, including one-hot encoding, standardisation, and feature engineering. The feature, “Rating_Reviews”, was introduced to capture the cumulative influence of ratings and reviews. The variable ‘Category’ was chosen as a target to discern different health contexts such as ‘Weight loss’ and ‘Medical’. Various machine learning and deep learning models were evaluated, from the baseline Random Forest Classifier to the sophisticated BERT model. The results highlighted the efficiency of transfer learning, especially BERT, which achieved an accuracy of approximately 90% after hyperparameter tuning. A final recommendation system was designed, which uses cosine similarity to rank apps based on their relevance to user queries.

BackgroundThe advances achieved in technology, medicine, and communications in the past decades have created an excellent scenario for the improvement and expansion of eHeath and mHealth in particular. Mobile phones, smartphones, and tablets are exceptional means for the application of mobile health, especially for those diseases and health conditions that are the deadliest worldwide.ObjectiveThe main aim of this paper was to compare the amount of research and the number of mobile apps dedicated to the diseases and conditions that are the leading causes of death according to the World Health Organization grouped by different income regions. These diseases and conditions were ischemic heart disease; stroke and other cerebrovascular diseases; lower respiratory infections; chronic obstructive pulmonary disease; diarrheal diseases; HIV/AIDS; trachea, bronchus, and lung cancers; malaria; and Alzheimer disease and other dementias.MethodsTwo reviews were conducted. In the first, the systems IEEE Xplore, Scopus, Web of Knowledge, and PubMed were used to perform a literature review of applications related to the mentioned diseases. The second was developed in the currently most important mobile phone apps stores: Google play, iTunes, BlackBerry World, and Windows Phone Apps+Games.ResultsSearch queries up to June 2013 located 371 papers and 557 apps related to the leading causes of death, and the following findings were obtained. Alzheimer disease and other dementias are included in the diseases with more apps, although it is not among the top 10 causes of death worldwide, whereas lower respiratory infections, the third leading cause of death, is one of the less researched and with fewer apps. Two diseases that are the first and second of low-income countries (lower respiratory infections and diarrheal diseases) have very little research and few commercial applications. HIV/AIDS, in the top 6 of low-income and middle-income zones, is one of the diseases with more research and applications, although it is not in the top 10 in high-income countries. Trachea, bronchus, and lung cancers are the third cause of death in high-income countries but are one of the least researched diseases with regard to apps.ConclusionsConcerning mobile apps, there is more work done in the commercial field than in the research field, although the distribution among the diseases is similar in both fields. In general, apps for common diseases of low- and middle-income countries are not as abundant as those for typical diseases of developed countries. Nevertheless, there are some exceptions such as HIV/AIDS, due to its important social conscience; and trachea, bronchus and lung cancers, which was totally unexpected.

Background: Road traffic accidents have been one of the main causes of death worldwide. However, most of the crashes are both predictable and preventable. With the widespread use, mobile phone is considered as a major road safety risk. However, varieties of apps have recently been developed for improving the road traffic health and safety. Despite an increasing trend of these apps, there is no comprehensive analysis of their features and no taxonomy or classification of them based on the traffic safety theories or frameworks. The aim of this study was to explore characteristics of available mobile apps on the road traffic health and safety and classify them based on the Haddon’s Matrix. Methods: To conduct this research, a comprehensive and systematic review of the mobile applications developed for the road traffic health and safety was carried out through using the qualitative content analysis. Google Play was searched through using a combination of the keywords to retrieve the road traffic apps. In order to extract the app features, their description was examined, and their content was analyzed. Then they were classified in four main categories including Road Traffic Health & safety, Road Traffic Training, Road traffic Navigation, and Other Road Traffic apps. Finally, the Haddon’s matrix was applied to analyze and classify those mobile apps residing in two main categories of the traffic health & safety, and the traffic training. Haddon’s matrix is a relevant framework for structured analyses of traffic injury events. In the Haddon’s matrix, the contributions of human, vehicle/equipment and environmental factors to the injuries in the three phases pre-crash, crash and post-crash has been presented. Results: In this study, 916 mobile apps met the inclusion criteria and were included in the final analysis.19 subcategories were identified for classifying the included apps based on their features and functionalities. In total, 620 of the mobile apps were grouped on the basis of Haddon’s matrix. About 61.76% of these apps were categorized as the traffic health & safety group. Among the subcategories of the traffic health & safety group, the highest number of apps (194) was the real-time traffic alerting apps. Behavior based feed-backing subcategory was ranked the second among the traffic health and safety apps with 75 apps. When the researchers applied factors and phases of Haddon’s framework for analyzing the apps, it was found that the Haddon’s factors have been taken into account in the apps grouped into two main categories of the apps including traffic training and traffic health & safety applications. The highest percentage of Haddon’s matrix factor considered for the intervention through the apps was related to the physical & social environment (23.69%) in the traffic training category. The features of 259 apps were classified in the Event/ Driving phase of Haddon’s matrix. Most of the apps in this phase was related to the real-time traffic alerting subgroup. These apps performed in real time while driving. The lowest number of the apps in driving phase was observed in the behavior based feed backing subcategory, which inform drivers about their driving behavior in real-time. Sixteen apps had features that categorized them in both Pre-event / Pre-driving and Event / Driving phases. In addition, 20 apps were found to have feature for intervening in both Event / Driving and Post-event / Post –driving phases. All 235 apps in the traffic training category were in pre-event/ pre-driving phase, which requires users to use these apps before driving. Conclusion: Applying Haddon’s matrix on analyzing and classifying the traffic health and safety apps revealed strengths and weaknesses of the existing related mobile apps in terms of the factors that must be considered for intervention toward prevention and reduction of road traffic accidents as the main public health issue.

Study question Does mobile applications (apps) developed for infertile couples have a holistic and utility characteristic? Summary answer Many mobil apps developed in the field of fertility have limited features and very few relevant, understandable, and holistic apps were present for infertile individuals. What is known already In an increasingly technological world, mobile apps offer promising alternatives to support couples’ fertility treatment. Infertile couples can use these apps with aim to manage lifestyle, learn about fertility, follow clinical appointments, receive psychological support, or to seek peer support. For increasing efficiency of this applications, holistic approach should be considered. Human is a whole with biological, physiological, psychological, social and spiritual dimensions and in fertility treatment process all dimensions are highly crucial. Thus, the purpose of this study was to evaluate the utility and holistic perspectives of apps developed for infertile couples by using “APPLICATIONS” mobile app scoring system. Study design, size, duration The study was a systematic evaluation of apps related to fertility and it was conducted via mobile applications platform. To determine relevant apps search was conducted on Google Play and Apple App Store by using “infertility”, “fertility”, “in vitro fertilization”, “IVF”, “sterility” and “fecundity” key words.The search was conducted twice in 1-year intervals, in February 2021 and March 2022. In total from 1917 reached apps; 1841 apps were excluded, and 76 eligible app were evaluated. Participants/materials, setting, methods Mobile apps were independently evaluated by two authors via using APPLICATIONS scoring system. The APPLICATIONS scoring system consists objective criteria such as comprehensiveness (holistic approach), fee, subscription, use of evidence-based information, purchasing, connection, advertising, text search area, interplatform suitability, use of images/shapes or videos, and app-specific features are objective scoring criteria and subjective criteria for instance subjective presentation and ease of navigation in the app. Consequently the maximum score of APPLICATIONS scoring system is 17. Main results and the role of chance The study reached 1917 applications from the search with its keywords. Seventy-six mobile applications meeting the inclusion criteria were analyzed according to the “APPLICATIONS” scoring system. Of the 76 applications in the analysis, 18 were available only in the Apple App Store (23%), 28 were available only in the Google Play Store (36%), and 30 were available on both platforms (41%). Among analyzed app, we determined that one app received 16 points out of maximum 17 points, and two apps received 15 points each. Also, the lowest scoring app earned 4 points. According to overall composite scores, evaluable apps in our study showed scores in the middle range of possible scores with an average score of 9 out of 17, indicating insufficient quality in mobile apps. Besides that, only six apps (%7.8) meet all holistic comprehensiveness criteria and received full scores. Of 76 mobile applications, 31.5% were developed by a healthcare professional, 31.5% by a software company, and 38% by an unknown entity. In addition, 48% of the applications evaluated were specific to infertile couples, 45% were specific to infertile women, and 7% were specific to infertile men. Limitations, reasons for caution Because of the nature of the digital environment, mobile apps are in a continuously changing and rapidly developing internet environment, and currently, newly developed mobile apps were not present in the analysis. This was the limitation of our study. Wider implications of the findings Using mobile health apps can help infertile couples to improve treatment outcomes. Healthcare professionals should be aware of the widespread use of apps and correct the faulty practice recommendations that patients may encounter, and couples should be guided to the correct practices. Trial registration number not applicable

Mobile apps have exploded in popularity, encouraging developers to provide content to the massive user base of the main app stores. Although there exist automated techniques that can classify user comments into various topics with high levels of precision, recent studies have shown that the top apps in the app stores do not have customer ratings that directly correlate with the app's success. This implies that no single requirements elicitation technique can cover the full depth required to produce a successful product and that applying alternative requirements gathering techniques can lead to success when these two are combined. Since user involvement has been found to be the most impactful contribution to project success, in this paper we will explore how the Wizard of Oz (WOz) technique and user reviews available in Google Play, can be integrated to produce a product that meets the demand of more stakeholders than either method alone. To compare the role of early interactive requirements specification and app reviews, we conducted two studies (i) a case study analysis on 13 mobile app development teams who used very early stages Requirements Engineering (RE) by applying WOz, and (ii) a study analyzing 40 (70, 592 reviews) similar mobile apps on Google Play. The results of both studies show that while each of WOz and app review analysis techniques can be applied to capture specific types of requirements, an integrated process including both methods would eliminate the communication gap between users and developers at early stages of the development process and mitigates the risk of requirements change in later stages.

Along with the increasing popularity of mobile devices, there exist severe security and privacy concerns for mobile apps. On Google Play, user reviews provide a unique understanding of security/privacy issues of mobile apps from users' perspective, and in fact they are valuable feedbacks from users by considering users' expectations. To best assist the end users, in this paper, we automatically learn the security/privacy related behaviors inferred from analysis on user reviews, which we call review-to-behavior fidelity. We design the system AUTOREB that automatically assesses the review-to-behavior fidelity of mobile apps. AUTOREB employs the state-of-the-art machine learning techniques to infer the relations between users' reviews and four categories of security-related behaviors. Moreover, it uses a crowdsourcing approach to automatically aggregate the security issues from review-level to app-level. To our knowledge, AUTOREB is the first work that explores the user review information and utilizes the review semantics to predict the risky behaviors at both review-level and app-level.

A Mobile Recommender System (MRS) is a system that provides personalized recommendations for mobile users. It solves the problem of information overload in a mobile environment with the support of a smart mobile device. MRS has three fundamental characteristics relevant to the mobile Internet: mobility, portability and wireless connectivity. MRS aims to generate accurate recommendations by utilizing detailed personal data and extracting user preferences. However, collecting and processing personal data may intrude user privacy. The privacy issues in MRS are more complex than traditional recommender system due to its specific characteristics and various personal data collection. Privacy protection in MRS is a crucial research topic, which is widely studied in the literature, but it still lacks a comprehensive survey to summarize its current status and indicate open research issues for further investigation. This paper reviews existing work in MRS in terms of privacy protection. Challenges and future research directions are discussed based on the literature survey.

The use of substandard and counterfeit medicines (SCM) leads to significant health and economic consequences, like treatment failure, rise of antimicrobial resistance, extra expenditures of individuals or households and serious adverse drug reactions including death. Our objective was to systematically search, identify and compare relevant available mobile applications (apps) for smartphones and tablets, which use could potentially affect clinical and public health outcomes. We carried out a systematic review of the literature in January 2020, including major medical databases, and app stores. We used the validated Mobile App Rating Scale (MARS) to assess the quality of apps, (1 worst score, 3 acceptable score, and 5 best score). We planned to evaluate the accuracy of the mobile apps to detect SCM. We retrieved 335 references through medical databases and 42 from Apple, Google stores and Google Scholar. We finally included two studies of the medical database, 25 apps (eight from the App Store, eight from Google Play, eight from both stores, and one from Google Scholar), and 16 websites. We only found one report on the accuracy of a mobile apps detecting SCMs. Most apps use the imprint, color or shape for pill identification, and only a few offer pill detection through photographs or bar code. The MARS mean score for the apps was 3.17 (acceptable), with a maximum of 4.9 and a minimum of 1.1. The ‘functionality’ dimension resulted in the highest mean score (3.4), while the ‘engagement’ and ‘information’ dimensions showed the lowest one (3.0). In conclusion, we found a remarkable evidence gap about the accuracy of mobile apps in detecting SCMs. However, mobile apps could potentially be useful to screen for SCM by assessing the physical characteristics of pills, although this should still be assessed in properly designed research studies.

In 2013, the global mobile app market was estimated at over US$50 billion and is expected to grow to $150 billion in the next two years. In this paper, we build a structural econometric model to quantify the vibrant platform competition between mobile (smartphone and tablet) apps on the Apple iOS and Google Android platforms and estimate consumer preferences toward different mobile app characteristics. We find that app demand increases with the in-app purchase option wherein a user can complete transactions within the app. On the contrary, app demand decreases with the in-app advertisement option where consumers are shown ads while they are engaging with the app. The direct effects on app revenue from the inclusion of an in-app purchase option and an in-app advertisement option are equivalent to offering a 28% price discount and increasing the price by 8%, respectively. We also find that a price discount strategy results in a greater increase of app demand in Google Play compared with Apple App Store, and app developers can maximize their revenue by providing a 50% discount on their paid apps. Using the estimated demand function, we find that mobile apps have enhanced consumer surplus by approximately $33.6 billion annually in the United States, and we discuss various implications for mobile marketing analytics, app pricing, and app design strategies.This paper was accepted by Alok Gupta, special issue on business analytics.

Reviewing the data security and privacy policies of mobile apps for depression

Content providers (CPs) increasingly deploy network infrastructures that oftentimes reach up to the wireless network edge, i.e. base stations or small cells. Hence, they are interested in optimizing resource allocation and relevant performance metrics for that infrastructure. On the other hand, mobile apps featuring streaming content (e.g. video, music) come with social-networking and content-sharing capabilities among users. These need to be taken into account in resource allocation since they decisively shape content demand. In this work, we introduce mathematical optimization problems about resource allocation at the wireless network edge, which obtain interesting twists when social-network diffusion is considered. Specifically, we consider, (i) the problem of content caching and user targeting through the recommender system of the app, with the goal to maximize the social diffusion effect of cached content, and (ii) the problem of user targeting through the mobile app recommender system, so that the available wireless bandwidth is utilized as efficiently as possible.