ASPG Menu
search

American Scientific Publishing Group

Research Feed

Found 3791 matches for "All Articles"

Exploratory Data Analysis of International Student Demographics: Trends, Insights, and Implications

This paper focuses on Exploratory Data Analysis of the data from the “International Student Demographics,” which is available on Kaggle and comprises data collected through the academic years, as well as total students, U.S. students, undergraduate, graduate, non-degree students, and OPT columns. In the given work, the author intends to provide a chronological overview of the demographic data of international students. Operations like handling missing values and outliers were done to prepare the data for a more elaborate analysis. All descriptive analyses during the study included time series plots and bar charts where time series was used to evidence key trends and distributions. The analyses of the research questions indicate that there has been growth in international student enrollment over the decades, particularly in undergraduate and OPT student categories, with influences from world events such as COVID-19 and the alteration of immigration policies. Country-wise contribution reveals that the maximum number of articles originated from East Asia and South and Central Asia, with a special focus on engineering, social sciences, and humanities. Solutions: The paper articulates the finality of trends affecting educational institutions and policymakers by focusing on the implications of international students’ demographics. Based on the findings above, future research directions are proposed to improve the findings and support evidence-based practice relating to international education.

groups
Sekar Kidambi Raju mail -
Marwa M. Eid mail
link https://doi.org/10.54216/JAIM.070204

Volume & Issue

Vol. Volume 7 / Iss. Issue 2

Details open_in_new

Proposed Two-Parameter Estimator for Estimating Linear Regression Model and Comparing It with Some Other Estimators

In this paper, a new two-parameter estimator was proposed to estimate the parameters of the linear regression model that has the ability to face the problem of Multicollinearity based on the previous information about the parameters to be estimated and this estimator was compared with the two-parameter estimator of the linear regression model of Kaciranlar and the two-parameter estimator of the linear regression model (Lokman et al. [1]) using the mean square error criterion (MSE) for each model by conducting Monte-Carlo simulation to study the behavior of the proposed estimator. It was concluded that the proposed method is better than the rest of the estimation methods because it achieved the lowest comparison criteria, and in the case of high Multicollinearity between the explanatory variables, the proposed method was very effective in solving this problem. Data representing (100) observations of the number of women with Irritable Bowel Syndrome (IBS) for the years (2020-2023) from the Karbala Holy Health Department were used, which represents the dependent variable (y) and a group of variables affecting the incidence of the disease, with nineteen variables. It was concluded that irritable bowel syndrome among women is decreasing, as the predictive values ​​according to the proposed method are appropriate for the estimated values ​​during the next five years.

groups
NoorAlzahraa Naeem Abd Ali mail -
Shrooq Abdul Redha Al Sabah mail
link https://doi.org/10.54216/PMTCS.040201

Volume & Issue

Vol. Volume 4 / Iss. Issue 2

Details open_in_new

Predicting Student Outcomes: Evaluating Regression Techniques in Educational Data

Student performance prediction is essential so that institutions can assist in identifying weak performers and initiate corrective measures. This research assesses different regression models by applying data from Kaggle, which involves data cleaning like managing missing values and scaling of the data, hence feature extraction, then model imposition and authenticity. The models followed are Linear Regression, SVR, MLPRegressor, Gradient Boosting, CatBoost, XGBoost, Random Forest, Extra Trees, Decision Tree and K-neighbors. The analysis shows that Linear Regression produced the best result as it has the lowest MSE score of 0.000521 and high accuracy regarding other measures, including RMSE, MAE, and R2. The results reveal that regression models can be used to predict students’ performance and be helpful to the various stakeholders in the system. The findings of this study will help develop required models for decision-making to improve students’ performance.

groups
Manish Kumar Singla mail -
Faris H. Rizk mail -
Mahmoud Elshabrawy Mohamed mail -
Ahmed Mohamed Zaki mail
link https://doi.org/10.54216/JAIM.070205

Volume & Issue

Vol. Volume 7 / Iss. Issue 2

Details open_in_new

Predictive Modeling of Global Educational Outcomes: A Comparative Analysis Using Machine Learning Regression Techniques

Education is central to social and economic development, making school enrollment and the quality of learning important global priorities. This study applies machine-learning regression to the World Educational Data dataset from Kaggle. After data exploration and preprocessing, eleven models are compared: Support Vector Regression (SVR), CatBoost, Random Forest, Extra Trees, XGBoost, a multilayer perceptron (MLP), Gradient Boosting, Decision Tree, K-Nearest Neighbors, Linear Regression, and a pipeline model. Performance is evaluated using MSE, RMSE, MAE, MBE, the correlation coefficient, R2, NSE, and WI. CatBoost reports the strongest correlation and coefficient of determination (both approximately 0.99999) with an MSE of 0.04024. The results illustrate the potential of machine-learning methods to support analysis and prediction of educational outcomes and to inform educational planning.

groups
Abdelhameed Ibrahim mail -
Abdelaziz A. Abdelhamid mail -
Ehab M. Almetwally mail
link https://doi.org/10.54216/JAIM.070206

Volume & Issue

Vol. Volume 7 / Iss. Issue 2

Details open_in_new

Securing DNS over HTTPS: A Machine Learning Study on Traffic Classification Using DoHBrw-2020

This study investigates machine-learning approaches for classifying secure DNS traffic and identifying threats associated with DNS over HTTPS (DoH). Using the DoHBrw-2020 dataset, we compare Decision Tree, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Naive Bayes, a multilayer perceptron (MLP), Gradient Boosting, and SVM with an RBF kernel. Models are evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F-score. The Decision Tree achieves the highest accuracy (99.65%) and strong results across the evaluation criteria. These findings demonstrate the potential of machine learning to improve DNS traffic security and support the development of real-time DoH threat detection. The results also inform the design of safer DNS solutions for contemporary information-security environments.

groups
Al-Seyday.T. Qenawy mail -
Hussein Alkattan mail -
Amany Khaled mail
link https://doi.org/10.54216/JAIM.070207

Volume & Issue

Vol. Volume 7 / Iss. Issue 2

Details open_in_new

Robust Estimation of Component Cost-Risk Indices in a CODLAG Naval Propulsion Digital Shadow

Maintenance-economic studies often convert condition indicators into apparent costs without verifying whether the resulting targets are independent of the model inputs. This can produce exceptionally high but operationally misleading accuracy. We formulate component-level estimation for a COmbined Diesel-eLectric And Gas (CODLAG) propulsion plant as a leakage-resistant digital-shadow problem. The primary dataset is the public UCI naval propulsion dataset: 11,934 steady-state observations generated by a calibrated numerical frigate simulator across nine speed settings, 51 compressor-health states, and 26 turbine-health states. Because it contains neither monetary maintenance records nor live vessel telemetry, the outputs are explicitly defined as dimensionless compressor and turbine cost-risk indices rather than currency costs. Health coefficients and every target-derived quantity are excluded from predictors. Models are tuned by three-fold grouped inner validation and assessed by five-fold outer validation in which all observations sharing a degradation pair remain in the same fold. Extra Trees achieved an aggregate cost-risk RMSE of 0.0222 (95% CI half-width 0.0049) and R2 of 0.9886 (95% CI half-width 0.0049) under nominal interpolation. However, combined noise, missingness, calibration bias, and channel loss reduced R2 to 0.6335 even after corruption-augmented training, while leave-one-speed-regime-out testing produced mean R2 = −0.379. Scenario-matched 90% split-conformal intervals retained 89.5–90.4% empirical coverage by widening from 0.032 nominally to 0.406 under severe corruption. An external NASA C-MAPSS FD001 benchmark yielded RMSE 16.97 cycles and R2 = 0.821 for remaining-useful-life prediction on 100 unseen engines. The study therefore supports robust condition-risk ranking within the simulated envelope, not direct monetary forecasting or autonomous maintenance scheduling. The principal contribution is a transparent validation protocol that exposes target leakage, simulator determinism, sensor-quality sensitivity, and the boundary between a digital shadow and a deployable digital twin.

groups
Samir Aboul Fotouh Saleh mail -
A. S. Tolba mail -
Shimaa Sami Selim mail -
MohamedElhoseny mail
link https://doi.org/10.54216/AJBOR.150107

Volume & Issue

Vol. Volume 15 / Iss. Issue 1

Details open_in_new

Numerical Proceduers for Computing the Exact Solutions to Systems of Ordinary Differential Equations

This paper presents a modified homotopy perturbation method (HPM), which aimed at solving systems of ordinary differential equations (ODEs). The MHPM, which combines the HPM, Laplace transform, and Padé approximants, offers an alternative approach to address the challenges associated with solving such problems. By employing this method, it becomes feasible to overcome these challenges and obtain a dependable approximation for the exact solution. The effectiveness and applicability of the proposed scheme are demonstrated through preliminary results derived from illustrative examples, all of which correspond to exact solutions.

groups
Nidal Anakira mail -
Osama Oqilat mail -
Adel Almalki mail -
Irianto Irianto mail -
Saad Meqdad mail -
Ala Amourah mail
link https://doi.org/10.54216/IJNS.250214

Volume & Issue

Vol. Volume 25 / Iss. Issue 2

Details open_in_new

Evaluating the Effectiveness of Physical Rehabilitation Exercises through RehabNet++ and Hybrid Optimization Techniques

This article focuses on improving the accuracy and efficiency of multimodal human motion analysis using advanced techniques. Initially, Generative Adversarial Networks (GANs) were used for skeletal enhancement, and then Contrast-Limited Adaptive Histogram Equalization (CLAHE) was applied on the enhanced images to check the quality Joint-level. Limb-level, Temporal, Statistical Features are effectively recovered from contrast enhancing images. Furthermore, with the selected optimal features acquired from PutterFish Customized Serval Optimizer (PFCSO), the RehabNet++ architecture that makes the human movement assessment has been trained. This PFCSO model has been developed based on the inspiration acquired from the Pufferfish Optimization Algorithm (POA) and the Serval Optimization algorithm (SOA), respectively. The RehabNet++ architecture includes an optimized Multilayer Perceptron (O-MLP), STR-ResNet architecture, Attention-based Convolutional Neural Networks and Transfer Learning. The O-MLP model has been formulated by optimizing the hidden layers of MLP using the PFCSO model. In addition, Grad-CAM visualization is included to provide a graphical description for model selection. A comparative study has been conducted to test the proposed deep learning algorithm against the original methods using the Kimore dataset. This analysis is implemented in PYTHON and is dedicated to multimodal human motion analysis.

groups
Osamah A. Altammami mail
link https://doi.org/10.54216/FPA.170112

Volume & Issue

Vol. Volume 17 / Iss. Issue 1

Details open_in_new

Biometrics Applied to Forensics Exploring New Frontiers in Criminal Identification

Different biological data may be used to identify people in this investigation. The system uses complex multimodal fusion, feature extraction, classification, template matching, adjustable thresholding, and more. A trustworthy multimodal feature vector (B) is created using the Multimodal Fusion Algorithm from voice, face, and fingerprint data. The key objectives are weighing, normalizing, and extracting characteristics. Complex feature extraction algorithms improve this vector and ensure its accuracy and reliability. Hamming distance is utilized in template matching for accuracy. Support vector machines to ensure classification accuracy. The adaptive threshold technique adjusts option limits based on the biology score mean and standard deviation when external conditions change. A thorough look at the research shows how algorithms operate together and how vital each aspect is for locating criminals. Change the multimodal fusion weights for optimum results. Thorough research using tables and photographs revealed that the fingerprint approach is optimal. Fast, simple, and precise technologies may enable new unlawful recognition tools. The adaptive thresholding algorithm's multiple adaptation steps allow the system to adjust to diverse study circumstances. The Multimodal Biometric Identification System is a cutting-edge leader in its area and provides a trustworthy, practical, and customizable research choice. This novel strategy is at the forefront of criminal recognition technology and has been supported by ablation research. It affects reliability, accuracy, and adaptability.

groups
Ajay Kushwaha mail -
Tushar Kumar Pandey mail -
B. Laxmi Kantha mail -
Prashant Kumar Shukla mail -
Sheo Kumar mail -
Rajesh Tiwari mail
link https://doi.org/10.54216/JCIM.150122

Volume & Issue

Vol. Volume 15 / Iss. Issue 1

Details open_in_new

The Challenge of Adversarial Attacks on AI-Driven Cybersecurity Systems

As AI is deployed increasingly in defensive systems, hostile assaults have increased. AI-driven defensive systems are vulnerable to attacks that exploit flaws. This article examines the approaches used to resist AI-based cybersecurity systems and their effects on security. This paper examines existing literature and case studies to demonstrate how attackers modify AI models. These include avoidance, poisoning, and data-driven assaults. It also considers data breaches, system failures, and unauthorized access if a hostile effort succeeds. The report recommends adversarial training, model testing, and input sanitization to address these issues. It also stresses the need for monitoring and updating AI algorithms to adapt to changing opponent tactics. This paper emphasizes the need to limit hostile strike threats using real-life examples and statistics. To defend AI-driven cybersecurity systems from complex threats, cybersecurity specialists, AI researchers, and policymakers must collaborate across domains. This article provides full guidance for cybersecurity and AI professionals. It describes the complex issues adversarial assaults create and proposes a flexible and robust architecture to safeguard AI-driven cybersecurity systems from emerging threats.

groups
M. N. V Kiranbabu mail -
A. Jeraldine Viji mail -
Amit Kumar Chandanan mail -
Vijay Birchha mail -
Tushar Kumar Pandey mail -
Sumit Kumar Sar mail
link https://doi.org/10.54216/JCIM.150123

Volume & Issue

Vol. Volume 15 / Iss. Issue 1

Details open_in_new