Popular Applications Research Articles

Analysis on Transfer Learning Models and Applications in Natural Language Processing

Assumptions have been established that many machine learning algorithms expect the training data and the testing data to share the same feature space or distribution. Thus, transfer learning (TL) rises due to the tolerance of the different feature spaces and the distribution of data. It is an optimization to improve performance from task to task. This paper includes the basic knowledge of transfer learning and summarizes some relevant experimental results of popular applications using transfer learning in the natural language processing (NLP) field. The mathematical definition of TL is briefly mentioned. After that, basic knowledge including the different categories of TL, and the comparison between TL and traditional machine learning models is introduced. Then, some applications which mainly focus on question answering, cyberbullying detection, and sentiment analysis will be presented. Other applications will also be briefly introduced such as Named Entity Recognition (NER), Intent Classification, and Cross-Lingual Learning, etc. For each application, this study provides reference on transfer learning models for related researches.

Neighborhood Characteristics Associated with Running in Metro Vancouver: A Preliminary Analysis.

Running can improve physical health and psychological wellbeing. However, the characteristics of conducive running environments are relatively unknown. This study determines neighborhood factors that attract running and explores how age and gender mediate built environment preferences. Spatial patterns of runners in Metro Vancouver were identified using crowdsourced fitness data from Strava, a popular application for tracking physical activities. The influence of socio-economic status (SES), green and/or blue space, and urbanicity on route popularity was assessed using a Generalized Linear Model (GLM). The influence of these neighborhood variables was also calculated for runners by age and gender. The results show high neighborhood SES, the presence of green and/or blue space, and high population density are associated with increased running activities in all age and gender groups. This study contributes a novel approach to understanding conducive running environments by demonstrating the utility of crowdsourced data in combination with data about urban environments. The patterns of this large group of runners can be used to inform planning for cities that promote running, as well as seek to encourage equal participation among different ages and genders.

From Theory to Code: Identifying Logical Flaws in Cryptographic Implementations in C/C++

Cryptographic protocols are often expected to be provably secure. However, this security guarantee often falls short in practice due to various implementation flaws. We propose a new paradigm called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">cryptographic program analysis (CPA)</i> which prescribes the use of program analysis to detect these implementation flaws at compile time. The principal insight of the CPA is that many of these flaws in cryptographic implementations can be mapped to the violation of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">meta-level properties</i> of implementations. A program property that is necessary to realize a cryptographic property is referred to as meta-level property. We show that violations of these meta-level properties can be identified at compile-time that can serve as sufficient evidence of the encompassing flaws. We investigated existing literature on cryptographic implementation flaws and derived 25 corresponding meta-level properties. To instantiate the abstract paradigm of CPA, we develop a specification language based on deterministic finite automaton (DFA) and show that most of the meta-level properties can be expressed in terms of our language. We then develop a tool called <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TaintCrypt</small> which uses static taint analysis to identify meta-level property violations of C/C++ cryptographic implementations at compile-time. We demonstrate the efficacy of <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TaintCrypt</small> by analyzing open-source C/C++ cryptographic libraries (e.g., OpenSSL) and observe that <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TaintCrypt</small> could have helped to avoid several high-profile flaws. We also evaluated <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TaintCrypt</small> on 5 popular applications and libraries, which generated new security insights. The experimental evaluation on large-scale projects indicates the scalability of our approach.

Poe in Richmond: Creating a Virtual Poe Collection

Poe in Richmond: Creating a Virtual Poe Collection

Dirty Metadata: Understanding A Threat to Online Privacy

People have a certain expectation of privacy when using widespread and trusted online services. This study surveys several popular web applications to understand if uploaded user images are handled securely using open source digital forensics tools and a custom software framework.

The benefits of simultaneous tDCS and working memory training on transfer outcomes: A systematic review and meta-analysis

BackgroundTranscranial direct current stimulation (tDCS) has shown potential as an effective aid to facilitate learning. A popular application of this technology has been in combination with working memory training (WMT) in order to enhance transfer effects to other cognitive measures after training. ObjectiveThis meta-analytic review aims to synthesize the existing literature on tDCS-enhanced WMT to quantify the extent to which tDCS can improve performance on transfer tasks after training. Furthermore, we were interested to evaluate the moderating effects of assessment time point (immediate post-test vs. follow-up) and transfer distance, i.e., the degree of similarity between transfer and training tasks. MethodsUsing robust variance estimation, we performed a systematic meta-analysis of all studies to date that compared WMT with tDCS to WMT with sham in healthy adults. All procedures conformed to PRISMA guidelines. ResultsAcross 265 transfer measures in 18 studies, we found a small positive net effect of tDCS on improving overall performance on transfer measures after WMT. These effects were sustained at follow-up, which ranged from 1 week to one year after training, with a median of 1 month. Additionally, although there were no significant differences as a function of transfer distance, effects were most pronounced for non-trained working memory tasks. ConclusionsThis review provides evidence that tDCS can be effective in promoting learning over and above WMT alone, and can durably improve performance on trained and untrained measures for weeks to months after the initial training and stimulation period. In particular, boosting performance on dissimilar working memory tasks may present the most promising target for tDCS-augmented WMT.

ZİHİNSEL SAĞLIK UYGULAMALARININ İKNA TEKNOLOJİLERİ BAĞLAMINDA BİR DEĞERLENDİRMESİ: VAKA İNCELEMESİ

Covid-19 pandemisi esnasında insanların internet tabanlı zihinsel sağlık uygulamalarına yönelmesi sonucu en popüler 10 meditasyon uygulamasının indirilme sayısı 2020’nin ilk çeyreğinde toplam 2 milyon artmıştır ve bu pazarın 2027 yılında 3,7 milyar dolara kadar büyüyeceği öngörülmektedir.&#x0D; Bu çalışmanın amacı, milyonlarca kullanıcının akıllı cihazlarına indirdiği zihinsel sağlık aplikasyonlarının bu başarısını anlamlandırmak adına kullanılan ikna stratejilerini ortaya çıkarmak ve karşılaştırmalı olarak analiz etmektir. Çalışmanın literatüründe bireylerin kriz zamanlarında yöneldiği zihinsel sağlık aplikasyonları, ikna ve kaptoloji bağlamında incelenmiştir. Çalışmanın araştırma bölümünde ise meditasyon aplikasyonlarında kullanılan ikna stratejilerini ortaya çıkarmak için vaka incelemesi yapılmış, amaçlı örneklem metoduyla seçilen 3 zihinsel sağlık aplikasyonu, Esgin’in; Cialdini’nin 6 ikna prensibini Fogg’un davranışsal modeline uygun sınıflandırdığı Davranış Kodlama Modeli Tablosu’na uyarlanarak incelenmiştir.&#x0D; Seçilen vakalar kaptoloji bağlamında incelendiğinde, 6 ikna prensibinden en az 5 tanesini kullandıkları gözlemlenmektedir. İkna kodlarını en başarılı şekilde kullanan aplikasyonun Meditopia olduğu sonucuna ulaşılmıştır.

Developing a procedure to produce large-scale topographic maps of open-pit mines using rotary wing unmanned aerial vehicles

Recently, the rapid development of unmanned aerial vehicle (UAV) has led to increasingly popular applications of this technology in the field of surveying and mapping. This article presents the result of developing a procedure to produce large-scale topographic maps for open-pit mines in Vietnam in which unmanned aerial vehicles (UAV) are used to collect data. With the goal of creating a procedure that ensures accuracy, safety, and efficiency for the production of topographic maps, the theories and official regulations for measuring and editing topographic maps in general and in mines in particular, as well as using drones in Vietnam have been applied. The procedure is specifically developed for the vertical takeoff and landing (rotary) UAVs. Experiments consisting of the flight altitude calculation, determining the appropriate take-off position, and image acquisition were conducted at two open-pit mines, namely Dong Da Mai coal mine and Long Son quarry to verify the procedure. The digital surface models (DSM) and orthophotos are generated from the acquired photos, georeferenced to the VN-2000 coordinate system using ground control points (GCP), and evaluated with checkpoints. In addition, the experiment included the investigation of the various combination of GCP networks to determine the optimal GCP number for each case. Experimental results show that the procedure is logical, correct, straightforward, safe, and effective. UAV resulted products, such as orthophotos, DSM, and topographic maps were accurate and met the requirements of all the national standards.

Impact of social influence on users' continuance intention toward sports and fitness applications.

The purpose of this paper is to explore how social influence (SI), which is disaggregated into subjective norms (SN), social image (SIM), and social identity (SID), predicts perceived usefulness (PU), perceived pleasure (PP), and continuance intention (CI) toward sports and fitness applications. The underlying context is the socialization and gamification of exercise during the Covid-19 pandemic. Based on the theory of SI and the technology acceptance model, a theoretical framework was built where PU and PP mediate the influence of SI on CI, and proposed hypotheses were tested. The responses of 296 Keep users (a popular sports and fitness application in China) to a questionnaire survey were analyzed. SN and SIM were found to have significant positive effects on SID; SID has significant positive effects on PU and PP; both PU and PP have significant positive effects on the CI of users; SID and PU positively and significantly mediate the relationship between SN/SIM and CI; PU positively and significantly mediates the SID-CI relationship. However, the role of PP in mediating the influence of SI on CI is non-significant. This paper deepens the current understanding of the mechanisms that influence the relationship between SI and CI under the context of socialization and gamification services.

Numerical study on the performance of the thermoelectric generator with different air-water mixtures

Thermoelectric generators (TEGs) have been widely used to convert thermal energy into electrical energy directly. Nowadays, the most popular TEG applications, such as vehicle heat recovery and geothermal power generation, typically utilize water as the cooling working fluid. Water sources, however, may not be available or extremely limited in some cases. Therefore, in this study, we proposed a new method of mixing water and air as the working fluid to save water. It is necessary to find out the suitable ratio of water to air, under which the performance loss could be minimized. A numerical model of TEG was built utilizing COMSOL Multiphysics to investigate the effect of water mole fraction (WMF, representing water to air ratio) on the performance of TEG, where the properties of the air-water mixture (AWM) could vary with WMF. In this study, 5 cases were considered at different flow velocities, different WMFs, and different combinations of air, water, and AWMs. The effects of these parameters on the voltage and the output power of TEGs were studied and analyzed. The TEG module could generate higher voltage and power output with increasing velocity and WMF. When air fixed on one side, only 0.2 of WMF on the other side could boost over 1.5 times of voltage and 3 times of power generation than only using air. Based on the simulation results, TEGs reach the best performance when the value of WMF is equal to 0.5 on both sides.

Gene Expression Analysis via Spatial Clustering and Evaluation Indexing

The density-based spatial clustering for applications with noise (DBSCAN) is one of the most popular applications of clustering in data mining, and it is used to identify useful patterns and interesting distributions in the underlying data. Aggregation methods for classifying nonlinear aggregated data. In particular, DNA methylations, gene expression. That show the differentially skewed by distance sites and grouped nonlinearly by cancer daisies and the change Situations for gene excretion on it. Under these conditions, DBSCAN is expected to have a desirable clustering feature i that can be used to show the results of the changes. This research reviews the DBSCAN and compares its performance with other algorithms, such as the traditional number of clustering, K-mean particle swarm optimization (PSO), and Grey–Wolf optimization (GWO). This method offers high performance for improvement. The DBSCAN algorithm also offers better results of clusters and gives better performance assessment according to the results shown in this study.

Rancang Bangun Kamus Digital Berbasis Chatbot Menggunakan Pendekatan Pattern Matching

The rapid development of technology today has had an impact on various fields of life, one of which is in the world of education. The application of information technology in the learning process at school or on campus has become commonplace. Smart devices such as smartphones connected to the internet can help the learning process and help encourage the communication process between educational stakeholders. In online learning, the process of communication and interaction between students and teachers switches through instant messaging providers. Taken from the 2019 digital report data issued by We Are Social Hotsuit, it is stated that the most popular instant messaging application in Indonesia is WhatsApp, where 83% of internet users in Indonesia are WhatsApp users, with this data making Indonesia one of the countries with the largest number of users in the world. Based on this, to support the performance of teachers in delivering learning materials, as well as provide faster access in searching for the required terms, this study built a digital chatbot-based dictionary that was embedded in the instant messaging application as the most popular message delivery medium in Indonesia. The process of building a chatbot-based digital dictionary in this study was carried out using the pattern matching method, which is a method included in the study of artificial intelligence that recognizes input from users to be matched with data stored in the database, and the appropriate response will be returned to the user. The subject content that is sampled in making this digital dictionary is physics. The result of this study is the existence of a chatbot-based digital dictionary that can assist students in finding definitions of various terms and formulas related to physics material, given by the teacher which students can access at any time as long as they are connected to the internet.

Avatars' Social Rhythms in Online Games Indicate Their Players' Depression.

Data sets on gameplay, called digital biomarkers, contain many characteristics of game players and are associated with mental health problems. In fact, an avatar's behavior during an online game is said to be related to its player's mental health. Based on this information, we estimated the depressive state of players based on their avatar's behavioral logs. There were 3,361 participants who were players of Pigg Party, a popular Japanese online game application. In April 2020, the players logged into the Pigg Party and answered a questionnaire on depression. In May 2020, of the 3,361 participants, 658 players again logged into the Pigg Party and answered the depression questionnaire. Their responses to the questionnaire and behavioral logs of Pigg Party for April and May comprised the data sets used in the study. The data set from April showed that individuals without depression were more likely to perform activities in a 24-hour cycle than individuals with depression. Furthermore, the popular model of the time series data set trained on the April data set predicted depression in the May data set with an accuracy of around 0.3 to 0.4. The results suggest that these models are difficult to predict for sparse time series data sets, such as the study's data set, and require refinement.

Focusing on your subject: Deep subject-aware image composition recommendation networks

Photo composition is one of the most important factors in the aesthetics of photographs. As a popular application, composition recommendation for a photo focusing on a specific subject has been ignored by recent deep-learning-based composition recommendation approaches. In this paper, we propose a subject-aware image composition recommendation method, SAC-Net, which takes an RGB image and a binary subject window mask as input, and returns good compositions as crops containing the subject. Our model first determines candidate scores for all possible coarse cropping windows. The crops with high candidate scores are selected and further refined by regressing their corner points to generate the output recommended cropping windows. The final scores of the refined crops are predicted by a final score regression module. Unlike existing methods that need to preset several cropping windows, our network is able to automatically regress cropping windows with arbitrary aspect ratios and sizes. We propose novel stability losses for maximizing smoothness when changing cropping windows along with view changes. Experimental results show that our method outperforms state-of-the-art methods not only on the subject-aware image composition recommendation task, but also for general purpose composition recommendation. We also have designed a multistage labeling scheme so that a large amount of ranked pairs can be produced economically. We use this scheme to propose the first subject-aware composition dataset SACD, which contains 2777 images, and more than 5 million composition ranked pairs. The SACD dataset is publicly available at https://cg.cs.tsinghua.edu.cn/SACD/.

Exploring the use of data compression for accelerating machine learning in the edge with remote virtual graphics processing units

SummaryInternet of Things (IoT) devices are usually low performance nodes connected by low bandwidth networks. To improve performance in such scenarios, some computations could be done at the edge of the network. However, edge devices may not have enough computing power to accelerate applications such as the popular machine learning ones. Using remote virtual graphics processing units (GPUs) can address this concern by accelerating applications leveraging a GPU installed in a remote device. However, this requires exchanging data with the remote GPU across the slow network. To address the problem with the slow network, the data to be exchanged with the remote GPU could be compressed. In this article, we explore the suitability of using data compression in the context of remote GPU virtualization frameworks in edge scenarios executing machine learning applications. We use popular machine learning applications to carry out such exploration. After characterizing the GPU data transfers of these applications, we analyze the usage of existing compression libraries for compressing those data transfers to/from the remote GPU. Our exploration shows that transferring compressed data becomes more beneficial as networks get slower, reducing transfer time by up to 10 times. Our analysis also reveals that efficient integration of compression into remote GPU virtualization frameworks is strongly required.

COMPARISON BETWEEN E-LEARNING APPLICATIONS DURING BLENDED LEARNING IN IRAQ

Blended Learning (BL) is the use of both offline and online methods for teaching and learning; it allows for a planned and deliberate embeddedness of online learning into the education system and tests its applicability alongside offline learning, which ensures adaptability to emerging challenges and the upskilling of the digital capabilities of Iraqi youth. This approach was a serious try to improve the education and E-learning and it indeed had a dramatic effect as it worked to promote the general well-being of learners in Iraq. This paper investigates and presents a comparison between the E-learning applications during BLA (in term of their features, flexibility, ease of use, advantages and disadvantages, and privacy and security). The outcome of the study was relied on a questionnaire made in various Iraqi universities. A total of 100 people participated in the questionnaire, divided into two groups: students and lecturers. The results showed that most lecturers and students prefer Google Meet. Google classroom and Zoom were the second popular applications whereas Microsoft Teams was the least choice for both groups.

Semantic models for IoT sensing to infer environment–wellness relationships

Every time an Internet of Things (IoT) solution is deployed, every time a smartphone owner connects her/his wireless device to a wearable activity-tracker, every time groups of citizens use geo-mapping applications to move around the city, choosing the least crowded path, data are produced and information have to be exchanged appropriately via APIs. Even if novel added-value IoT-based applications appear on the market with increasing speed, true semantic interoperability is far from being achieved, thus limiting the large-scale exploitation, the scalability and the time-to-market of novel apps. Currently, connecting different data prosumers with multiple data sources is still hampered by the lack of standardized and sustainable solutions, especially due to the significant heterogeneity of IoT platforms. In such a landscape, ontologies come to the rescue, thanks to their formal semantics, knowledge representation formats, and shared vocabularies. In this paper we examine, from an ontological perspective, how to describe environmental sensing and wellness monitoring, two of the most popular application cases of Mobile Crowd Sensing (MCS) and IoT, respectively. To this purpose, an ontology of sensor-agnostic APIs is proposed, along with a set of MCS-dedicated ontology modules (and the supporting platform), leveraging on standard and reusable domain ontologies. Moreover, it will be shown how to properly combine the proposed ontologies in order to support complex functionalities based on inference rules addressing the environment–wellness relationships. Finally, specific semantic modeling patterns suitable for typical IoT and MCS scenarios will be discussed.

Application of indocyanine green (ICG)-guided surgery in clinical practice: lesson to learn from other organs-an overview on clinical applications and future perspectives.

Indocyanine green (ICG) fluorescence-guided surgery is a modality of intra-operative navigation that might support the surgeon with enhanced visualization of anatomical structures in real time. Over the last years, it has emerged as one of the most promising and rapidly developing technical innovations in surgery. The most popular current clinical applications include fluorescence cholangiography, bowel anastomotic perfusion assessment, fluorescence-guided lymphography for sentinel lymph-node identification and guided lymphadenectomy and the possible use in oncological surgery for the identification and localization of tumors and the diagnosis and treatment of peritoneal carcinosis. This paper provides an overview of the multiple fields of applications of ICG fluorescence-guided surgery in visceral and oncological surgery, discussing indications summarizing most recent and significative available literature and giving technical notes of use.

Fatty acids and nutritional components of the seed oil from Wangmo red ball Camellia oleifera grown in the low-heat valley of Guizhou, China

Wangmo red ball Camellia oleifera is the main Camellia species cultivated for oil in the low-heat valley of Guizhou, China. In this study, we evaluated the comprehensive nutritional value of Wangmo C. oleifera seed oil through fatty acid and nutritional component analyses. Twenty excellent Wangmo C. oleifera plants with stable yield and disease resistance were selected from the Camellia oleifera germplasm resource garden in the low-heat valley site of Guizhou University. The unit crown yield, fatty acid content of the seed oil, fatty acid composition and functional nutrients were determined, and the oil quality was comprehensively evaluated using principal component analysis. The fatty acid content of C. oleifera seed oil was 35.03–53.47%, suggesting likely popularization and wide application prospects. The fatty acids included SFAs, MUFAs and PUFAs, and the oleic acid content was 80%, indicating a highly stable and nutritious oil. The oil was also rich in carotenoids, polyphenols, flavonoids, β-sitosterol, squalene and α-Ve, with average content of 7.404 mg/kg, 16.062 mg/kg, 0.401 g/100 g, 265.087 mg/kg, 129.315 mg/kg and 21.505 mg/100 g, respectively. However, the correlations among the nutritional indices were weak. PCA showed that germplasms GH7, GH43, GH28, GH8 and GH31 exhibited the top five nutritional qualities. The rankings in this study provide data for identifying excellent Wangmo C. oleifera plants with high nutritional quality. Additionally, this study provides a valuable reference for the research and development of high-end edible oil and a theoretical basis for the development of economic forest species in low-heat valley areas across the world.

High scintillation yield and fast response to alpha particles from thin perovskite films deposited by pulsed laser deposition

Although the field of solar cells is the most popular application of perovskite materials, their use in radiation detection applications is emerging. The success of perovskites as radiation detectors rests partly on the same material properties that have led to successful optoelectronics applications, meaning that other specific properties, like large stopping power, high mobility lifetime product, fast response, and large bulk resistance, play a role. In this respect, inorganic perovskites are attracting a lot of attention as scintillator materials with performances sensitive to material shape (single crystals, nanocrystals, and thin films) and growth methods. In this study, we report on the morphological, structural, and optical response of thin CsPbBr3-based perovskite films, deposited by Pulsed Laser Deposition (PLD) and post-growth annealed at 350°C in air, following excitation by different particle sources. The annealing treatment resulted in a prompt structural refinement, grain growth, and oxygen bonding to the Pb phase together with an enrichment of the surface in chemo-adsorbed oxygen probably due to Cs–O interactions, as evidenced by X-ray photoelectron spectroscopy. The film behavior under 2 MeV H+ ion beam irradiation at different fluences was analyzed together with its scintillation properties following an interaction with α particles from an Am-241 radioactive source demonstrating a very fast response for an inorganic material (∼5ns) and a photoelectron yield of about ∼47% with respect to a commercial CsI:Tl scintillator.