Search

Android is the most popular operating system of the latest mobile smart devices. With this operating system, many Android applications have been developed and become an essential part of our daily lives. Unfortunately, different kinds of Android malware have also been generated with these applications' endless stream and somehow installed during the API calls, permission granted and extra packages installation and badly affected the system security rules to harm the system. Therefore, it is compulsory to detect and classify the android malware to save the user's privacy to avoid maximum damages. Many research has already been developed on the different techniques related to android malware detection and classification. In this work, we present AMDDLmodel a deep learning technique that consists of a convolutional neural network. This model works based on different parameters, filter sizes, number of epochs, learning rates, and layers to detect and classify the android malware. The Drebin dataset consisting of 215 features was used for this model evaluation. The model shows an accuracy value of 99.92%. The other statistical values are precision, recall, and F1-score. AMDDLmodel introduces innovative deep learning for Android malware detection, enhancing accuracy and practical user security through inventive feature engineering and comprehensive performance evaluation. The AMDDLmodel shows the highest accuracy values as compared to the existing techniques. [ABSTRACT FROM AUTHOR]

To explore the visual experience characteristics and influencing factors of college students' visual attention intervention in the interface of mobile learning platform by using eye-tracking technology, and to summarize and summarize the visual experience pattern of platform interface design and its design inspiration. Methods: Using the head-mounted eye-tracking technology, 28 images from 6 groups of typical elements in the interface layout of CGTN learning platform were selected as the test samples, and the eye-movement indexes of the subjects browsing the interface were recorded. Results: There were significant differences in the attention time, number of times of attention, visual attention rate and visual recall rate of different areas and topics of the interface (P < 0. 001). Conclusion: In the platform interface design, the analysis of the factors influencing visual attention can be found that people's attention and visual experience is mainly influenced by color, text, and typography, and secondary areas and layout also play an important role in visual communication. The color and text areas in the interface design, as well as the innovative design of typography can effectively enhance the visual attention of college students and better communicate the information of the platform. [ABSTRACT FROM AUTHOR]

Children typically prefer to attend to social stimuli (e.g. faces, smiles) over non-social stimuli (e.g. natural scene, household objects). This preference for social stimuli is believed to be an essential building block for later social skills and healthy social development. Preference for social stimuli are typically measured using either passive viewing or instrumental choice paradigms, but not both. Since these paradigms likely tap into different mechanisms, the current study addresses this gap by administering both of these paradigms on an overlapping sample. In this study, we use a preferential looking task and an instrumental choice task to measure preference for social stimuli in 3–9 year old typically developing children. Children spent longer looking at social stimuli in the preferential looking task but did not show a similar preference for social rewards on the instrumental choice task. Task performance in these two paradigms were not correlated. Social skills were found to be positively related to the preference for social rewards on the choice task. This study points to putatively different mechanisms underlying the preference for social stimuli, and highlights the importance of choice of paradigms in measuring this construct. [ABSTRACT FROM AUTHOR]

Mobile sensing data processed using machine learning models can passively and remotely assess mental health symptoms from the context of patients' lives. Prior work has trained models using data from single longitudinal studies, collected from demographically homogeneous populations, over short time periods, using a single data collection platform or mobile application. The generalizability of model performance across studies has not been assessed. This study presents a first analysis to understand if models trained using combined longitudinal study data to predict mental health symptoms generalize across current publicly available data. We combined data from the CrossCheck (individuals living with schizophrenia) and StudentLife (university students) studies. In addition to assessing generalizability, we explored if personalizing models to align mobile sensing data, and oversampling less-represented severe symptoms, improved model performance. Leave-one-subject-out cross-validation (LOSO-CV) results were reported. Two symptoms (sleep quality and stress) had similar question-response structures across studies and were used as outcomes to explore cross-dataset prediction. Models trained with combined data were more likely to be predictive (significant improvement over predicting training data mean) than models trained with single-study data. Expected model performance improved if the distance between training and validation feature distributions decreased using combined versus single-study data. Personalization aligned each LOSO-CV participant with training data, but only improved predicting CrossCheck stress. Oversampling significantly improved severe symptom classification sensitivity and positive predictive value, but decreased model specificity. Taken together, these results show that machine learning models trained on combined longitudinal study data may generalize across heterogeneous datasets. We encourage researchers to disseminate collected de-identified mobile sensing and mental health symptom data, and further standardize data types collected across studies to enable better assessment of model generalizability. [ABSTRACT FROM AUTHOR]

Large-scale ecological sampling can be difficult and costly, especially for organisms that are too small to be easily identified in a natural environment by eye. Typically, these microscopic floral and fauna are sampled by collecting substrates from nature and then separating organisms from substrates in the laboratory. In many cases, diverse organisms can be identified to the species-level using molecular barcodes. To facilitate large-scale ecological sampling of microscopic organisms, we used a geographic data-collection platform for mobile devices called Fulcrum that streamlines the organization of geospatial sampling data, substrate photographs, and environmental data at natural sampling sites. These sampling data are then linked to organism isolation data from the laboratory. Here, we describe the easyFulcrum R package, which can be used to clean, process, and visualize ecological field sampling and isolation data exported from the Fulcrum mobile application. We developed this package for wild nematode sampling, but it can be used with other organisms. The advantages of using Fulcrum combined with easyFulcrum are (1) the elimination of transcription errors by replacing manual data entry and/or spreadsheets with a mobile application, (2) the ability to clean, process, and visualize sampling data using a standardized set of functions in the R software environment, and (3) the ability to join disparate data to each other, including environmental data from the field and the molecularly defined identities of individual specimens isolated from samples. [ABSTRACT FROM AUTHOR]

As the Internet retail industry continues to rise, more and more consumers choose to shop online, especially Chinese consumers. Using consumer behavior data left on the Internet to predict repurchase behavior is of great significance for companies to achieve precision marketing. This paper proposes an improved deep forest model, and the interactive behavior characteristics of users and goods are added into the original feature model to predict the repurchase behavior of e-commerce consumers. Based on the Alibaba mobile e-commerce platform data set, first construct a feature engineering that includes user characteristics, product characteristics, and interactive behavior characteristics. And then use our proposed model to make predictions. Experiments show that the model's overall performance with increased interactive behavior features is better and has higher accuracy. Compared with the existing prediction models, the improved deep forest model has certain advantages, which not only improves the prediction accuracy but also reduces the cost of training time. [ABSTRACT FROM AUTHOR]

We demonstrate the possibility of conducting synchronous, repeated, multi-game economic decision-making experiments with hundreds of subjects in-person or remotely with live streaming using entirely mobile platforms. Our experiment provides important proof-of-concept that such experiments are not only possible, but yield recognizable results as well as new insights, blurring the line between laboratory and field experiments. Specifically, our findings from 8 different experimental economics games and tasks replicate existing results from traditional laboratory experiments despite the fact that subjects play those games/task in a specific order and regardless of whether the experiment was conducted in person or remotely. We further leverage our large subject population to study the effect of large (N = 100) versus small (N = 10) group sizes on behavior in three of the scalable games that we study. While our results are largely consistent with existing findings for small groups, increases in group size are shown to matter for the robustness of those findings. [ABSTRACT FROM AUTHOR]

Photo-identification (photo-id) is a method used in field studies by biologists to monitor animals according to their density, movement patterns and behavior, with the aim of predicting and preventing ecological risks. However, these methods can introduce subjectivity when manually classifying an individual animal, creating uncertainty or inaccuracy in the data as a result of the human criteria involved. One of the main objectives in photo-id is to implement an automated mechanism that is free of biases, portable, and easy to use. The main aim of this work is to develop an autonomous and portable photo-id system through the optimization of image classification algorithms that have high statistical dependence, with the goal of classifying dorsal fin images of the blue whale through offline information processing on a mobile platform. The new proposed methodology is based on the Scale Invariant Feature Transform (SIFT) that, in conjunction with statistical discriminators such as the variance and the standard deviation, fits the extracted data and selects the closest pixels that comprise the edges of the dorsal fin of the blue whale. In this way, we ensure the elimination of the most common external factors that could affect the quality of the image, thus avoiding the elimination of relevant sections of the dorsal fin. The photo-id method presented in this work has been developed using blue whale images collected off the coast of Baja California Sur. The results shown have qualitatively and quantitatively validated the method in terms of its sensitivity, specificity and accuracy on the Jetson Tegra TK1 mobile platform. The solution optimizes classic SIFT, balancing the results obtained with the computational cost, provides a more economical form of processing and obtains a portable system that could be beneficial for field studies through mobile platforms, making it available to scientists, government and the general public. [ABSTRACT FROM AUTHOR]

Android is the most widely used mobile operating system (OS). A large number of third-party Android application (app) markets have emerged. The absence of third-party market regulation has prompted research institutions to propose different malware detection techniques. However, due to improvements of malware itself and Android system, it is difficult to design a detection method that can efficiently and effectively detect malicious apps for a long time. Meanwhile, adopting more features will increase the complexity of the model and the computational cost of the system. Permissions play a vital role in the security of the Android apps. Term Frequency—Inverse Document Frequency (TF-IDF) is used to assess the importance of a word for a file set in a corpus. The static analysis method does not need to run the app. It can efficiently and accurately extract the permissions from an app. Based on this cognition and perspective, in this paper, a new static detection method based on TF-IDF and Machine Learning is proposed. The system permissions are extracted in Android application package's (Apk's) manifest file. TF-IDF algorithm is used to calculate the permission value (PV) of each permission and the sensitivity value of apk (SVOA) of each app. The SVOA and the number of the used permissions are learned and tested by machine learning. 6070 benign apps and 9419 malware are used to evaluate the proposed approach. The experiment results show that only use dangerous permissions or the number of used permissions can't accurately distinguish whether an app is malicious or benign. For malware detection, the proposed approach achieve up to 99.5% accuracy and the learning and training time only needs 0.05s. For malware families detection, the accuracy is 99.6%. The results indicate that the method for unknown/new sample's detection accuracy is 92.71%. Compared against other state-of-the-art approaches, the proposed approach is more effective by detecting malware and malware families. [ABSTRACT FROM AUTHOR]

Introduction: Crowdsourcing, a participatory approach to solicit ideas from a large group of diverse individuals, provides an opportunity to nurture youth participation in HIV self-testing service design. The objective of this study was to describe the responses to a crowdsourcing contest aimed at soliciting ideas on how to promote HIV self-testing (HIVST) among young people in Nigeria. Methods: The "4 Youth by Youth" HIV Self-Testing Crowdsourcing contest was an online and off-line contest that asked Nigerian youth (10–24 years old) for ideas in response to the following: How might we promote HIVST among young people in Nigeria? All data were collected and analyzed between October 2018, and June 2019. Ideas and perceptions generated from the crowdsourcing contest were qualitatively analyzed using thematic content analysis. Specifically, four reviewers analyzed whether the ideas generated were desirable (appealing to young people), feasible (easy to implement) and impactful (will significantly influence HIVST uptake among young people). Results: A total of 903 entries were received in response to the contest prompt. Participants submitted entries in various forms: online form (39.7%), offline Dropbox (44.6%), email (6.1%) and WhatsApp (9.7%). Of the total entries, 85% (n = 769/903) entries were eligible and were scored as having either high, moderate or low level of feasibility, impact and desirability, on a 3-point Likert scale. A significant portion of the entries were given a score of 3 for feasibility (4.9%), desirability (7.1%), impact (3.0%) or a total overall score of 7 or more (8.2%). The three main themes that emerged from the entries include:1) Peer-to-peer distribution and leveraging on existing infrastructures 2) Youth-Oriented Branding of the HIVST Kit 3) Mobile platforms and social media technology. Conclusion: The "4 Youth by Youth" Self-Testing contest engaged a broad audience of young people to generate ideas and perspectives on how to promote HIVST. This process informed the development of youth innovated implementation strategies to increase uptake of HIVST among adolescents and youth at risk for HIV. [ABSTRACT FROM AUTHOR]

A hand-made, radiolucent, custom-designed device having a mobile and a non-mobile platforms was used to objectively quantify the in vivo cranial tibial translation, in order to assess the functional status of cranial-cruciate-ligament (CrCL) in dogs. The hypothesis was that changes in CrCL integrity would result in detectable changes in tibial translation. To validate the diagnostic method, data from injured (PA, n = 32), contralateral (CO, n = 32) and healthy stifles (HE, n = 32) were compared. Normalized tibial translation (ΔN) of each stifle was measured in medio-lateral radiographic projection obtained before and during standard thrust force application, in PA (43.59±12.97%), CO (20.32±6.69%) and HE (12.22±3.77%). Comparing PA with HE and CO (ΔN cut-off value: 29.73%), diagnosis could be issued with very high probability. Comparing HE with CO (ΔN cut-off value: 14.80%), high performance was obtained. The translator device could be a useful tool to objectively quantify the in vivo tibial translation in dogs with CrCL rupture, before surgery and during post-operatory follow-up. [ABSTRACT FROM AUTHOR]

Traditional search tasks have taught us much about vision and attention. Recently, several groups have begun to use multiple-target search to explore more complex and temporally extended “foraging” behaviour. Many of these new foraging tasks, however, maintain the simplified 2D displays and response demands associated with traditional, single-target visual search. In this respect, they may fail to capture important aspects of real-world search or foraging behaviour. In the current paper, we present a serious game for mobile platforms, developed in Unity3D, in which human participants play the role of an animal foraging for food in a simulated 3D environment. Game settings can be adjusted, so that, for example, custom target and distractor items can be uploaded, and task parameters, such as the number of target categories or target/distractor ratio are all easy to modify. We are also making the Unity3D project available, so that further modifications can also be made. We demonstrate how the app can be used to address specific research questions by conducting two human foraging experiments. Our results indicate that in this 3D environment, a standard feature/conjunction manipulation does not lead to a reduction in foraging runs, as it is known to do in simple, 2D foraging tasks. [ABSTRACT FROM AUTHOR]

Mobile cloud computing, which integrates the cloud computing techniques into the mobile environment, is regarded as one of the enabler technologies for 5G mobile wireless networks. There are many sporadic spare resources distributed within various devices in the networks, which can be used to support mobile cloud applications. However, these devices, with only a few spare resources, cannot support some resource-intensive mobile applications alone. If some of them cooperate with each other and share their resources, then they can support many applications. In this paper, we propose a resource cooperative provision mode referred to as "Crowd-funding", which is designed to aggregate the distributed devices together as the resource provider of mobile applications. Moreover, to facilitate high-efficiency resource management via dynamic resource allocation, different resource providers should be selected to form a stable resource coalition for different requirements. Thus, considering different requirements, we propose two different resource aggregation models for coalition formation. Finally, we may allocate the revenues based on their attributions according to the concept of the "Shapley value" to enable a more impartial revenue share among the cooperators. It is shown that a dynamic and flexible resource-management method can be developed based on the proposed Crowd-funding model, relying on the spare resources in the network. [ABSTRACT FROM AUTHOR]

The purpose of this study was to develop and validate a multimedia software application for mobile platforms to assist in the teaching and learning process of design and construction of a skin flap. Traditional training in surgery is based on learning by doing. Initially, the use of cadavers and animal models appeared to be a valid alternative for training. However, many conflicts with these training models prompted progression to synthetic and virtual reality models. Fifty volunteer fifth- and sixth-year medical students completed a pretest and were randomly allocated into two groups of 25 students each. The control group was exposed for 5 minutes to a standard text-based print article, while the test group used multimedia software describing how to fashion a rhomboid flap. Each group then performed a cutaneous flap on a training bench model while being evaluated by three blinded BSPS (Brazilian Society of Plastic Surgery) board-certified surgeons using the OSATS (Objective Structured Assessment of Technical Skill) protocol and answered a post-test. The text-based group was then tested again using the software. The computer-assisted learning (CAL) group had superior performance as confirmed by checklist scores (p<0.002), overall global assessment (p = 0.017) and post-test results (p<0.001). All participants ranked the multimedia method as the best study tool. CAL learners exhibited better subjective and objective performance when fashioning rhomboid flaps as compared to those taught with standard print material. These findings indicate that students preferred to learn using the multimedia method. [ABSTRACT FROM AUTHOR]