ASPG Menu
search

American Scientific Publishing Group

Research Feed

Found 3791 matches for "All Articles"

Multimodal Cognitive Workload Recognition in Human-Computer Interaction Using Biosignals and Interaction Traces

The process of recognizing cognitive workload requires reliable methods because researchers need to use both physiological indicators and interaction traces while facing challenges of limited data and inconsistent feature sets. The paper develops a multimodal fusion system which uses weight-based reliability assessment to identify three different workload levels from Cognitive Lab data which is publicly accessible. The subset which focuses on workload includes N-Back and mental subtraction tasks together with electroen-cephalography and functional near-infrared spectroscopy and electrocardiography and electrodermal activity and respiration and accelerometry and gaze descriptors and keyboard-mouse interaction indicators. The method conducts separate training for every modality through multidimensional variable reduction which enables gradient-boosted learners to make predictions about branch reliability based on their validation log-loss scores and combine posterior probabilities using normalized reliability weights. The design preserves distinct modality structures while controlling unpredictable branch effects. The study tests different approaches by evaluating single-modality learners against three methods which include direct early fusion and uniform late fusion and the proposed fusion rule. The proposed model achieves its best performance with 0.842 accuracy and 0.836 macro F1-score on the three-class workload task which includes the medium-load category that presents the greatest challenge to differentiate. The research results from class-wise and sensitivity assessments showed that interaction traces together with fNIRS features produced the smallest improvement to the system, and moderate reliability temperatures showed the highest stability in fusion pro-file performance. The feature attribution demonstrates specific emphasis on how cursor-velocity variability together with fNIRS oxygenation slope and EEG theta-band power and fixation-duration statistics and phasic electrodermal activity function as primary discriminative signals. The research findings demonstrate that multiple modal workload estimation needs to be improved through branch-specific modeling which should use decision fusion based on reliability as its foundation model and work through adaptive learning systems which have to handle rising cognitive requirements.

groups
Andino Maseleno mail -
Kharchenko Raisa mail -
Rahul Chauhan mail
link https://doi.org/10.54216/JCHCI.100203

Volume & Issue

Vol. Volume 10 / Iss. Issue 2

Details open_in_new

Task-Conditioned Early Prediction of Navigation Failure in Information Architecture Evaluation

The interaction logs which researchers collected during their information-architecture evaluation process contain detailed proof which shows how users select between successful and unsuccessful navigation routes. The predictive signal displays its initial appearance during task execution yet users exhibit different navigation patterns depending on their current task and interface they are using. The researchers of this study developed an early navigation failure prediction system which uses public interaction data to create task-specific prefix classification models. The study analyzes data from an open dataset which includes 180 participants completing 1800 tasks across six testing conditions that evaluate tree testing and highfidelity prototype navigation. A prefix-structural encoder works together with a regularized task-conditioned logistic model which predicts success based on the first k navigation actions. The researchers assessed model performance through participant-specific validation using three different machine learning techniques which included random forest, extra trees, and gradient boosting. The optimal configuration achieved 0.7833 accuracy, 0.7513 balanced accuracy, 0.8350 F1-score, and 0.7949 ROC–AUC performance at k = 3. The horizon analysis demonstration shows that predictive signals become accessible after users complete their first three actions. The ablation study proves that task conditioning functions as an essential component. The study results demonstrate that early trace analytics enable quick identification of navigation failures in information-architecture research while providing a useful method for customized assessment during usability testing.

groups
Kharchenko Raisa mail -
Rahul Chauhan mail -
Andino Maseleno mail
link https://doi.org/10.54216/JCHCI.110101

Volume & Issue

Vol. Volume 11 / Iss. Issue 1

Details open_in_new

A Systematic Literature Review on the Integration of Computer Vision and IoT Technologies for Enhancing Voter Verification Accuracy in Electoral Systems

The rapid evolution of digital technologies has transformed how societies manage sensitive information and authenticate identity in critical systems. Within the domain of cybersecurity and artificial intelligence (AI), the integration of computer vision and Internet of Things (IoT) technologies has emerged as a promising approach to improving real-time data verification and process automation. This systematic literature review examines how computer vision and IoT technologies can be jointly leveraged to enhance voter verification accuracy in electoral systems. Following the PRISMA 2020 guidelines, the review systematically searched four academic databases, identifying 351 initial studies. After rigorous screening based on predefined inclusion and exclusion criteria, 15 studies were selected for comprehensive analysis. The findings reveal three major themes: (1) emerging technical architectures combining biometric authentication with blockchain-based verification, (2) performance outcomes demonstrating high accuracy rates (97–100%) in controlled environments, and (3) persistent challenges in scalability, real-world deployment, and security against sophisticated AI-enabled attacks such as deepfakes. While the PRISMA process was conducted in full, the limited scope of the project, compressed timeline, and restricted access to paywalled articles likely influenced the depth and completeness of the synthesis. Nevertheless, the review provides structured insight into current implementation approaches, technical methods, and research gaps, with particular relevance to contexts like Uzbekistan where recent OSCE ODIHR election observation reports have documented systemic weaknesses in voter verification and turnout reporting.

groups
Angela Choi mail -
Eugene Q. Castro mail
link https://doi.org/10.54216/JCHCI.110102

Volume & Issue

Vol. Volume 11 / Iss. Issue 1

Details open_in_new

An Interaction-Centric Wireless Multimodal Fusion Model for Cognitive State Recognition in Computer Interfaces

Wireless human-computer interaction increasingly depends on distributed sensing, yet adaptive computer interfaces are still commonly modelled from isolated evidence streams. This paper presents an interaction-centric wireless multimodal fusion model for recognizing cognitive state during computer-based task execution. The model integrates wearable physiology, ocular behaviour, compact neurophysiological summaries, and direct interaction evidence obtained from the task interface, then adjusts each sensing channel through a reliability term that reflects wireless degradation. The experimental workflow follows a public stress-resilience human-computer interaction protocol involving synchronized task phases and computer interaction logs. The analysis shows that interaction variables such as task error, response latency, and click activity are among the strongest indicators of cognitive state and complement physiological information in a meaningful way. The results support the design of adaptive computer interfaces that respond not only to what the user is doing on the screen, but also to how reliably the supporting wireless sensing infrastructure is functioning.

groups
Khaled Sh. Gaber mail -
Mahmoud Elshabrawy Mohamed mail
link https://doi.org/10.54216/JCHCI.110103

Volume & Issue

Vol. Volume 11 / Iss. Issue 1

Details open_in_new

Cross-Modal Memory Support in Visually Demanding Environments: A Controlled Study of Haptic Pulses and Spatial Audio Cues for Reducing Prospective Memory Failures During Multitasking

When people are immersed in a visually demanding task, the attentional resources required to monitor the environment for cues that should trigger a remembered intention are frequently captured by the primary task, causing prospective memory failures that range from the inconvenient to the safety-critical. This problem is pervasive in modern work environments in which digital interfaces compete continuously for visual attention, yet the overwhelming majority of reminder and notification systems rely on the same visual channel that is already congested. This paper reports a controlled user study examining whether carefully designed haptic and spatial audio cues can compensate for this visual saturation and restore prospective memory performance without substantially increasing cognitive burden. Thirty-two participants completed a counterbalanced within-subjects protocol in which they performed primary cognitive tasks—document editing on a virtual desktop and navigating in a driving simulation—while managing a set of time-critical intentions delivered through four reminder conditions: visual-only, haptic-only, spatial audio only, and the combined haptic-plus-audio channel. The study measures prospective memory hit rate, task-switching errors, cue response latency, and multidimensional subjective workload across both scenarios and all four conditions. Results consistently favour the combined modality, which produces substantially fewer memory failures and lower reported workload than any single channel, while individual differences in baseline workload predict the magnitude of benefit from non-visual cueing. These findings carry direct implications for the design of ambient notification systems in high-demand professional and safety-critical environments.

groups
A. Nithya mail -
B. Chitra mail -
V. Sathya Preiya mail
link https://doi.org/10.54216/JCHCI.110104

Volume & Issue

Vol. Volume 11 / Iss. Issue 1

Details open_in_new

Measuring Visibility and Usability Features in Mobile Application Interface Design

Mobile application usability is often discussed after deployment through user reviews or task testing, but many visible design problems can be measured earlier from the interface itself. This paper presents a feature-based framework for quantifying mobile interface visibility, usability, and accessibility risk from screen-level design properties. The study defines a Mobile Interface Visibility–Usability Quality score using observable measures such as primary-action salience, visual density, tap-target adequacy, label completeness, contrast proxy, navigation depth, whitespace, and clutter. The analysis uses a structured extract following public Rico and UICrit-style mobile UI data, where screenshots, hierarchy information, and designer critique concepts support data-driven assessment. The results show that usability quality is not determined by a single visual property. Screens with strong contrast may still be difficult to use if feature discoverability is weak, and screens with many functions may remain usable when hierarchy and labels are clear. The paper contributes a measurement protocol, design risk taxonomy, empirical score analysis, and practical remediation loop for mobile app teams seeking objective evidence before user-facing release.

groups
Wadhah Abdullah mail -
Aygul Z. Ibatova mail
link https://doi.org/10.54216/JCHCI.110105

Volume & Issue

Vol. Volume 11 / Iss. Issue 1

Details open_in_new

From Signals to Action: Explainable AI for Engagement-Responsive Instructional Support in Digital Higher Education

Artificial intelligence (AI) is increasingly used to monitor learning processes in higher education, but many analytics pipelines stop at prediction and provide limited support for instructional action. This study develops an explainable AI framework that uses behavioral and contextual data from digital learning environments to generate tailored instructional support. Using the 480-record xAPI-Edu-Data dataset, the framework constructs an engagement index and support profiles, predicts multiclass performance, and allocates actions through rules. Three classifiers are compared using stratified cross-validation; Random Forest performs best overall. On the hold-out sample, it achieves 0.8021 accuracy, 0.8204 macro-precision, 0.8010 macro-recall, 0.8084 macro-F1, and 0.9140 macro-AUC. Student absence, the composite engagement index, gender, student–guardian relationship, support profile, and digital-resource access are the most influential predictors. The decision layer assigns students to attendance-first intervention, adaptive engagement support, family-engagement reinforcement, structured progression coaching, or challenge-and-extend pathways. The resulting workflow connects explainable AI with actionable instructional decision support for digital higher education.

groups
Andino Maseleno mail -
Meinhaj Hussain mail -
Aygul Z. Ibatova mail
link https://doi.org/10.54216/IJAIET.050101

Volume & Issue

Vol. Volume 5 / Iss. Issue 1

Details open_in_new

Early Detection of Student Dropout Risk in Higher Education through Optimized Machine Learning

Student retention in higher education institutions is a critical problem that causes academic and financial challenges to individual students, schools, and entire countries. The present study compares five machine-learning classifiers—Linear Discriminant Analysis (LDA), K-Nearest Neighbours (KNN), Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting (GBM)—using data for 4,424 students from the Realinho et al. (2022) dataset. The dataset contains demographic, socioeconomic, macroeconomic, and academic-performance information from a Portuguese higher-education institution. Mutual-information feature selection reduces the 22-dimensional feature space to the 12 features with the highest discriminative power. Five-fold stratified cross-validation shows that the SVM with a radial-basis-function kernel achieves the best overall performance, with 97.1% accuracy and an F1-score of 0.954; all five models achieve an AUC greater than 0.981. The importance analysis reveals that four measures of academic success from the first two semesters account for 87.6% of the Random Forest feature importance, driven by the number of curricular units passed during the second semester (importance = 0.335). Socioeconomic, demographic, and macroeconomic features together contribute less than 13%. These findings provide empirical evidence about student-retention risk factors and inform the design of higher-education early-warning systems.

groups
Aa Hubur mail -
Aygul Z. Ibatova mail
link https://doi.org/10.54216/IJAIET.040102

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Analytic Trait Scoring of English Language Learner Essays Using Fine-Tuned DeBERTa-v3 on the ELLIPSE Corpus

Automated Essay Scoring (AES) technologies have been extensively researched for holistic, topic-specific scoring, but their use to predict multiple analytic writing quality traits of English Language Learner (ELL) student essays has received less attention. This research contributes to this knowledge gap by systematically investigating multi-trait AES on the ELLIPSE corpus (Learning Agency Lab, 2022), a publicly accessible dataset of 6,482 argumentative essays written by grades 8–12 ELLs and rated by human raters on six analytic traits: cohesion, syntax, vocabulary, phraseology, grammar, and conventions. We experiment with five models: Ridge regression, Support Vector Regression (SVR) with a radial basis function (RBF) kernel, Random Forest, fine-tuned BERT-base-uncased, and fine-tuned DeBERTa-v3-base. The mean Quadratic Weighted Kappa (QWK) across six traits is highest for DeBERTa-v3 (0.726), a 26.5-point improvement over the Ridge baseline (0.461) and a 6-point improvement over BERT (0.666). Phraseology is the most difficult trait to score automatically (DeBERTa QWK = 0.701) and cohesion the easiest (DeBERTa QWK = 0.742). Analysis of inter-trait correlations reveals high co-variation between vocabulary and phraseology (r = 0.79), which may reflect common linguistic skills that can be leveraged by multi-task learning. This research sets a replicable baseline for multi-trait AES on the ELLIPSE corpus and suggests that phraseology scoring is the most urgent area for future architectural innovation.

groups
Andino Maseleno mail -
Meinhaj Hussain mail
link https://doi.org/10.54216/IJAIET.040103

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new

Educational Value Formation Around Intelligent Learning Tools: Student Readiness, Usage Archetypes, and Support Pathways in Higher Education

Modern higher education campuses now use intelligent learning tools as standard educational resources, yet students’ learning results depend on their understanding of these tools and their implementation in academic work. The study analyzes how students prepare to use educational tools while investigating the connection between their preparedness and their judgment of educational benefits. The study uses an open student-perception dataset to conduct empirical research that includes developing constructs, profiling readiness, creating predictive models, and establishing pathways. It introduces two measurement methods: source breadth, which measures how students acquire knowledge about intelligent tools through different information channels, and an advantage score, which presents perceived benefits for educational activities. The three-profile segmentation method shows that different groups in the sample display distinct levels of preparedness and value assessment. The Random Forest model demonstrates superior performance because it achieves the highest accuracy among all tested models in the predictive stage. The selected model exhibits an accuracy rate of 0.789, a precision rate of 0.714, a recall rate of 1.000, an F1 score of 0.833, and an area under the receiver operating characteristic curve of 0.806 in hold-out evaluation. The analysis of variable importance indicates that AI knowledge, grade-point average, information breadth, and profile membership serve as the main factors that explain the results. The final stage transforms analytical results into distinct educational pathways that focus on developing essential literacy skills, implementing structured curriculum materials, providing support for governance matters, and enabling advanced collaborative learning. The results demonstrate that the educational benefits of intelligent tools depend more on students’ preparedness to use them than on their initial exposure to the tools.

groups
Marina Sagatovna Abdurashidova mail -
Muhammad Eid Balbaa mail
link https: // doi. org/ 10. 54216/ IJAIET. 040104

Volume & Issue

Vol. Volume 4 / Iss. Issue 1

Details open_in_new