Search Results (33)

This paper presents a novel framework for optimizing multimodal large language model (MLLM) inference through task offloading and resource allocation in UAV-assisted satellite edge computing (SEC) networks. MLLMs leverage transformer architectures to integrate heterogeneous data modalities for IoT applications, particularly real-time monitoring in remote areas. However, cloud computing dependency introduces latency, bandwidth, and privacy challenges, while IoT device limitations require efficient distributed computing solutions. SEC, utilizing low-earth orbit (LEO) satellites and unmanned aerial vehicles (UAVs), extends mobile edge computing to provide ubiquitous computational resources for remote IoTDs. We formulate the joint optimization of MLLM task offloading and resource allocation as a mixed-integer nonlinear programming (MINLP) problem, minimizing latency and energy consumption while optimizing offloading decisions, power allocation, and UAV trajectories. To address the dynamic SEC environment characterized by satellite mobility, we propose an action-decoupled soft actor–critic (AD-SAC) algorithm with discrete–continuous hybrid action spaces. The simulation results demonstrate that our approach significantly outperforms conventional deep reinforcement learning methods in convergence and system cost reduction compared to baseline algorithms. Full article

This paper introduces an innovative mobile-assisted language-learning (MALL) system that harnesses deep learning technology to analyze pronunciation patterns and deliver real-time, personalized feedback. Drawing inspiration from how the human brain processes speech through neural pathways, our system analyzes multiple speech features using spectrograms, mel-frequency cepstral coefficients (MFCCs), and formant frequencies in a manner that mirrors the auditory cortex’s interpretation of sound. The core of our approach utilizes a convolutional neural network (CNN) to classify pronunciation patterns from user-recorded speech. To enhance the assessment accuracy and provide nuanced feedback, we integrated a fuzzy inference system (FIS) that helps learners identify and correct specific pronunciation errors. The experimental results demonstrate that our multi-feature model achieved 82.41% to 90.52% accuracies in accent classification across diverse linguistic contexts. The user testing revealed statistically significant improvements in pronunciation skills, where learners showed a 5–20% enhancement in accuracy after using the system. The proposed MALL system offers a portable, accessible solution for language learners while establishing a foundation for future research in multilingual functionality and mobile platform optimization. By combining advanced speech analysis with intuitive feedback mechanisms, this system addresses a critical challenge in language acquisition and promotes more effective self-directed learning. Full article

Mobile-assisted language learning (MALL) has emerged as a powerful tool for language education, offering flexibility, multimedia integration, and personalized learning experiences. Despite its growing adoption, most studies have focused on user perceptions and learning outcomes, with limited attention given to systematically evaluating the design, content, and pedagogical efficacy of mobile language learning apps (MLLAs). To address this gap in the effective design of MALL tools, this study developed an evaluation framework by integrating and refining elements from three established models in the field. The framework is organized into four dimensions: background and characteristics, app design, app content, and app pedagogy. It incorporates objective criteria alongside a standardized scoring system (0–2) to ensure consistent and systematic evaluations. The resulting framework provides researchers and educators with a tool to analyze and compare MLLAs based on their alignment with effective teaching and learning principles. This study contributes to the advancement of MALL app evaluation, supporting their development and improving teaching practices and learner outcomes. Full article

Programming has gained significant importance worldwide as societies increasingly rely on computer application systems. To support novices in learning various programming languages, we have developed the Programming Learning Assistant System (PLAS). It offers several types of exercise problems with different learning goals and levels for step-by-step self-study. As a personal answer platform in PLAS, we have implemented a web application using Node.js and EJS for Java and Python programming. Recently, the Flutter framework with Dart programming has become popular, enabling developers to build applications for mobile, web, and desktop environments from a single codebase. Thus, we have extended PLAS by implementing the Flutter environment with Visual Studio Code to support it. Additionally, we have developed an image-based user interface (UI) testing tool to verify student source code by comparing its generated UI image with the standard one using the ORB and SIFT algorithms in OpenCV. For efficient distribution to students, we have generated Docker images of the answer platform, Flutter environment, and image-based UI testing tool. In this paper, we present the implementation of a web-based answer platform for the Flutter Programming Learning Assistant System (FPLAS) by integrating three Docker images using Docker Compose. Additionally, to capture UI images automatically, an Nginx web application server is adopted with its Docker image. For evaluations, we asked 10 graduate students at Okayama University, Japan, to install the answer platform on their PCs and solve five exercise problems. All the students successfully completed the problems, which confirms the validity and effectiveness of the proposed system. Full article

The growing significance of English in global communication has heightened interest in sustainable English as a Foreign Language (EFL) instruction. Regrettably, sustainable EFL education has received insufficient attention, especially in the Indonesian context. To address this issue, recent conceptual research by Gayatri et al. has put forth a practical recommendation, namely the Sustainable Blended Global Discussion (SBGD) method, which combines classroom and online learning to engage students in global discussions with foreigners, promoting critical thinking. Despite being constructed on a solid foundation of theory and Indonesian context, the success of SBGD remains to be demonstrated. Hence, a multiple case study was conducted to examine the adaptation of SBGD in EFL classes at different universities. The EFL teachers were interviewed to explore the method’s advantages and challenges. Questionnaires were employed to study the perception and the critical thinking skills of 57 students, with some of them also being interviewed. Results showed that students indicated positive perception of the implementation of SBGD (M = 4.02 and M = 4.05). Additionally, students demonstrated a higher level of critical thinking skills through the teacher’s SBGD implementation in teaching and learning. Furthermore, greater student engagement, improved English language skills, and improved critical thinking were all observed; however, improvements like smaller group discussions, more facilitator involvement, and institutional supports were needed. This study is significant in addressing challenges and recommending the SBGD method as a solution for implementing online technologies in under-resourced contexts, specifically Indonesian higher education. The findings contribute to the literature on blended teaching and digital tools for second language education, with broader implications for similar educational settings. Full article

Stroke is a leading cause of long-term disability worldwide. With the advancements in sensor technologies and data availability, artificial intelligence (AI) holds the promise of improving the amount, quality and efficiency of care and enhancing the precision of stroke rehabilitation. We aimed to identify and characterize the existing research on AI applications in stroke recovery and rehabilitation of adults, including categories of application and progression of technologies over time. Data were collected from peer-reviewed articles across various electronic databases up to January 2024. Insights were extracted using AI-enhanced multi-method, data-driven techniques, including clustering of themes and topics. This scoping review summarizes outcomes from 704 studies. Four common themes (impairment, assisted intervention, prediction and imaging, and neuroscience) were identified, in which time-linked patterns emerged. The impairment theme revealed a focus on motor function, gait and mobility, while the assisted intervention theme included applications of robotic and brain–computer interface (BCI) techniques. AI applications progressed over time, starting from conceptualization and then expanding to a broader range of techniques in supervised learning, artificial neural networks (ANN), natural language processing (NLP) and more. Applications focused on upper limb rehabilitation were reviewed in more detail, with machine learning (ML), deep learning techniques and sensors such as inertial measurement units (IMU) used for upper limb and functional movement analysis. AI applications have potential to facilitate tailored therapeutic delivery, thereby contributing to the optimization of rehabilitation outcomes and promoting sustained recovery from rehabilitation to real-world settings. Full article

Nowadays, with the common use of smartphones in daily lives, mobile applications have become popular around the world, which will lead to a rise in Flutter framework. Developed by Google, Flutter with Dart programming provides a cross-platform development environment to create visually appealing and responsive user interfaces across mobile, web, and desktop platforms using a single codebase. However, due to time and staff limitations, the Flutter/Dart programming course is not included in curricula, even in IT departments in universities. Therefore, independent learning environments for students are essential to meet this growing popularity. Previously, we have developed programming learning assistant system (PLAS) as a web-browser-based self-learning platform for novice students. PLAS offers various types of exercise problems designed to cultivate programming skills step-by-step through a lot of code reading and code writing practices. Among them, one particular type is the code modification problem (CMP), which asks to modify the given source code to satisfy the new specifications. CMP is expected to be solved by novices with little effort if they have knowledge of other programming languages. Thus, PLAS with CMP will be an excellent platform for independent learning. In this paper, we present PLAS with CMP for the independent learning of Flutter/Dart programming. To improve the readability of the source code by students, we provided rich comments on grammar or behaviors. Besides, the code can be downloaded so that students can check and run it on an IDE. For evaluations, we generated 38 CMP instances for basic and multimedia/storage topics in Flutter/Dart programming and assigned them to 21 master students at Okayama University, Japan, who have never studied it. The results confirm the validity of the proposal. Full article

The communicative approach to language learning, a teaching method commonly used in second language (L2) classrooms, places little to no emphasis on pronunciation training. As a result, mobile-assisted pronunciation training (MAPT) platforms provide an alternative to classroom-based pronunciation training. To date, there have been several meta-analyses and systematic reviews of mobile-assisted language learning (MALL) studies, but only a few of these meta-analyses have concentrated on pronunciation. To better understand MAPT’s impact on L2 learners’ perceptions and production of targeted pronunciation features, this study conducted a systematic review of the MAPT literature following PRISMA 2020 guidelines. Potential mobile-assisted articles were identified through searches of the ERIC, Educational Full Text, Linguistics and Language Behavior Abstract, MLI International, and Scopus databases and specific journal searches. Criteria for article inclusion in this study included the following: the article must be a peer-reviewed empirical or quasi-empirical research study using both experimental and control groups to assess the impact of pronunciation training. Pronunciation training must have been conducted via MALL or MAPT technologies, and the studies must have been published between 2014 and 2024. A total of 232 papers were identified; however, only ten articles with a total of 524 participants met the established criteria. Data pertaining to the participants used in the study (nationality and education level), the MPAT applications and platforms used, the pronunciation features targeted, the concentration on perception and/or production of these features, and the methods used for training and assessments were collected and discussed. Effect sizes using Cohen’s d were also calculated for each study. The findings of this review reveal that only two of the articles assessed the impact of MAPT on L2 learners’ perceptions of targeted features, with results indicating that the use of MPAT did not significantly improve L2 learners’ abilities to perceive segmental features. In terms of production, all ten articles assessed MPAT’s impact on L2 learners’ production of the targeted features. The results of these assessments varied greatly, with some studies indicating a significant and large effect of MAPT and others citing non-significant gains and negligible effect sizes. The variation in these results, in addition to differences in the types of participants, the targeted pronunciation features, and MAPT apps and platforms used, makes it difficult to conclude that MAPT has a significant impact on L2 learners’ production. Furthermore, the selected studies’ concentration on mostly segmental features (i.e., phoneme and word pronunciation) is likely to have had only a limited impact on participants’ intelligibility. This paper provides suggestions for further MAPT research, including increased emphasis on suprasegmental features and perception assessments, to further our understanding of the effectiveness of MAPT for pronunciation training. Full article

Sign language is widely used to facilitate the communication process between deaf people and their surrounding environment. Sign language, like most other languages, is considered a complex language which cannot be mastered easily. Thus, technology can be used as an assistive tool to solve the difficulties and challenges that deaf people face during interactions with society. In this study, an automatic bidirectional translation framework for Arabic Sign Language (ArSL) is designed to assist both deaf and ordinary people to communicate and express themselves easily. Two main modules were intended to translate Arabic sign images into text by utilizing different transfer learning models and to translate the input text into Arabic sign images. A prototype was implemented based on the proposed framework by using several pre-trained convolutional neural network (CNN)-based deep learning models, including the DenseNet121, ResNet152, MobileNetV2, Xception, InceptionV3, NASNetLarge, VGG19, and VGG16 models. A fuzzy string matching score method, as a novel concept, was employed to translate the input text from ordinary people into appropriate sign language images. The dataset was constructed with specific criteria to obtain 7030 images for 14 classes captured from both deaf and ordinary people locally. The prototype was developed to conduct the experiments on the collected ArSL dataset using the utilized CNN deep learning models. The experimental results were evaluated using standard measurement metrics such as accuracy, precision, recall, and F1-score. The performance and efficiency of the ArSL prototype were assessed using a test set of an 80:20 splitting procedure, obtaining accuracy results from the highest to the lowest rates with average classification time in seconds for each utilized model, including (VGG16, 98.65%, 72.5), (MobileNetV2, 98.51%, 100.19), (VGG19, 98.22%, 77.16), (DenseNet121, 98.15%, 80.44), (Xception, 96.44%, 72.54), (NASNetLarge, 96.23%, 84.96), (InceptionV3, 94.31%, 76.98), and (ResNet152, 47.23%, 98.51). The fuzzy matching score is mathematically validated by computing the distance between the input and associative dictionary words. The study results showed the prototype’s ability to successfully translate Arabic sign images into Arabic text and vice versa, with the highest accuracy. This study proves the ability to develop a robust and efficient real-time bidirectional ArSL translation system using deep learning models and the fuzzy string matching score method. Full article

In the last fifteen years, an immense expansion has been witnessed in mobile app usage and production. The intense competition in the tech sector and also the rapidly and constantly evolving user requirements have led to increased burden on mobile app creators. Nowadays, fulfilling users’ expectations cannot be readily achieved and new and unconventional approaches are needed to permit an interested crowd of users to contribute in the introduction of creative mobile apps. Indeed, users and developers of mobile apps are the most influential candidates to engage in any of the requirements engineering activities. The place where both can best be found is on Twitter, one of the most widely used social media platforms. More interestingly, Twitter is considered as a fertile ground for textual content generated by the crowd that can assist in building robust predictive classification models using machine learning (ML) and natural language processing (NLP) techniques. Therefore, in this study, we have built two classification models that can identify mobile apps users and developers using tweets. A thorough empirical comparison of different feature extraction techniques and machine learning classification algorithms were experimented with to find the best-performing mobile app user and developer classifiers. The results revealed that for mobile app user classification, the highest accuracy achieved was ≈0.86, produced via logistic regression (LR) using Term Frequency Inverse Document Frequency (TF-IDF) with N-gram (unigram, bigram and trigram), and the highest precision was ≈0.86, produced via LR using Bag-of-Words (BOW) with N-gram (unigram and bigram). On the other hand, for mobile app developer classification, the highest accuracy achieved was ≈0.87, produced by random forest (RF) using BOW with N-gram (unigram and bigram), and the highest precision was ≈0.88, produced by multi-layer perception neural network (MLP NN) using BERTweet for feature extraction. According to the results, we believe that the developed classification models are efficient and can assist in identifying mobile app users and developers from tweets. Moreover, we envision that our models can be harnessed as a crowd selection approach for crowdsourcing requirements engineering activities to enhance and design inventive and satisfying mobile apps. Full article

In this paper, we present a sustainable approach for addressing the language skills gap among EU citizens, which significantly hinders their mobility across the EU and their participation in education, in training, as well as in youth programmes. Our approach is based on the sustainable design of the OpenLang Network platform, which provides an open and collaborative online learning environment for language learners and teachers across Europe, and addresses the limitations of existing computer-assisted language learning approaches. The OpenLang Network platform is bringing together educators and Erasmus+ mobility participants to improve their language skills and cultural knowledge. To this end, the OpenLang Network platform offers a collection of multilingual Open Educational Resources and language learning services. The paper presents the results from the user evaluation of the platform, which has been conducted with members of its community of language teachers and learners. A mixed methods approach has been adopted in order to collect and analyse both qualitative and quantitative data from users about the sustainable design of the OpenLang Network platform, as well as to measure the user satisfaction levels of the platform’s language learning services. According to the user evaluation results, the platform offers a sustainable online environment and a positive user experience for language learning. The user evaluation has also helped us identify a set of best practices and challenges associated with the long-term sustainability of an online language learning community. Full article

Assistive robots are tools that people living with upper body disabilities can leverage to autonomously perform Activities of Daily Living (ADL). Unfortunately, conventional control methods still rely on low-dimensional, easy-to-implement interfaces such as joysticks that tend to be unintuitive and cumbersome to use. In contrast, vocal commands may represent a viable and intuitive alternative. This work represents an important step toward providing a viable vocal interface for people living with upper limb disabilities by proposing a novel lightweight vocal command recognition system. The proposed model leverages the MobileNet2 architecture, augmenting it with a novel approach to the self-attention mechanism, achieving a new state-of-the-art performance for Keyword Spotting (KWS) on the Google Speech Commands Dataset (GSCD). Moreover, this work presents a new dataset, referred to as the French Speech Commands Dataset (FSCD), comprising 4963 vocal command utterances. Using the GSCD as the source, we used Transfer Learning (TL) to adapt the model to this cross-language task. TL has been shown to significantly improve the model performance on the FSCD. The viability of the proposed approach is further demonstrated through real-life control of a robotic arm by four healthy participants using both the proposed vocal interface and a joystick. Full article

The present study intended to examine the relationship between perceived teacher support, students’ ICT self-efficacy, and online English academic engagement in the blended learning setting, especially in mobile-assisted foreign language instruction contexts. A sample of 960 Chinese undergraduate and postgraduate students was recruited to participate in the online questionnaire. SPSS version 24.0 was used for descriptive, correlation, independent samples t-test, and mediation analysis of the three variables. The results showed that: (1) there is a significant correlation between perceived teacher support, students’ ICT self-efficacy, and online English academic engagement; (2) students’ ICT self-efficacy partially mediates the relationship between perceived teacher support and student online English academic engagement; (3) students’ ICT self-efficacies differed by sex and level of education, but not by major; (4) students’ sense of self-competence in ICT self-efficacy has a significant positive influence on engagement with online English learning. The findings reveal that students’ ICT self-efficacy positively impacts students’ online English learning, and perceived teacher support also affects students’ learning engagement. School administrators should encourage teachers to focus on students’ online self-efficacy, especially the sense of environmental control. Implications and further directions for future research are presented at the end. Full article

The outbreak of COVID-19 has impacted conventional educational practice in universities worldwide. Chinese universities are no exception. WeChat, a social application widely used in China, has been considered a viable tool for language education. However, the perspectives of Chinese university students and English language teachers in terms of using WeChat for English vocabulary learning and teaching during the pandemic remain unclear. The aim of the present study was twofold: First, it explored Chinese university students’ and language teachers’ opinions of adopting a self-developed WeChat-assisted lexical-learning program (the WALL program) during COVID-19. Second, it gathered their evaluations of the WALL program. To achieve the aim, two sets of semi-structured interviews were used to gather qualitative data about five students’ and three English language teachers’ perceptions at a university in northern China. The results first revealed that the eight participants showed overwhelming opinions in support of adopting the program for vocabulary learning and teaching during the pandemic. In addition, it received mostly positive evaluations. However, the program had two main drawbacks: distracting learning environments and uncertain learning effects. The present study then made recommendations for future WeChat-based language learning and teaching programs. The findings are expected to provide pedagogical insights for tertiary educational institutions, practitioners, and students in the chosen context in order to deal with the future design and implementation of sound MALL-based approaches. Full article

In recent decades, the rapid development of artificial intelligence (AI) technology has led to the increasing use of AI speaking apps in foreign language learning. This research investigates the impact of social network-based interaction on students’ English speaking practice with the assistance of AI speaking apps in China. During the summer vacation, 70 students from different Chinese universities and majors were recruited for the experiment. They were required to practice speaking skills with AI apps for five weeks and were divided into two groups. Participants in the experimental group were encouraged to engage in various interactive activities when practicing speaking with AI apps, while those in the control group were asked to use AI speaking apps without interaction. Data were collected through questionnaires and semi-structured interviews as well as pre-and post-tests. The results indicated that students generally held positive attitudes towards interactive activities when using AI apps to practice their spoken English. The finding also showed that social network-based interaction can effectively improve learners’ speaking skills in the AI context. This study contributes to the research on the implementation and promotion of AI speaking apps with social networking and extends the previous studies on network-based interaction to the AI-assisted learning environment. An investigation of interactions based on Chinese social network-based platforms such as WeChat can be further applied to other social networking platforms such as Facebook or WhatsApp in different cultural contexts for AI-assisted speaking practice. Full article