KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 10 of 57

Some of the metrics are blocked by your
consent settings
Low-cost Multispectral Acquisition Device Coupled with Machine Learning for Detecting Adulteration of Honey
(2026-07-15)
Boodnon, Wutthiphong
;
Lunvongsa, Thayanont
;
Suntisakoonwong, Phanchay
;
Sitorus, Agustami
;
Lapcharoensuk, Ravipat
Honey is a natural sweetener created by honeybees from the nectar of flowers. Honey's extensive health benefits have led to its widespread use across multiple industries. Honey adulteration with inferior substances undermines its quality, reducing natural nutrients and antioxidants, and diminishing its health benefits. This study aimed to study the possibility of detection of honey adulteration with a low-cost multispectral device coupled with machine learning. The adulterated honey came from deliberate adulteration with cane syrup in the 1 to 90% range. Spectral data was collected for pure honey and the adulterated honey samples at the wavelengths of 610, 680, 730, 760, 810, and 860 nm. The detection models for distinguishing pure and adulterated honey were developed by Linear Discriminant Analysis (LDA), Partial Least Squares Discriminant Analysis (PLS-DA), C-Support Vector Machine (C-SVM), and K-Nearest Neighbors (KNN). All models achieved high accuracy between 0.91 and 0.98 and maintained balanced precision and recall metrics. This study serves as a guideline for developing a low-cost portable honey authentication device that is practical for real-world applications.
Some of the metrics are blocked by your
consent settings
DDoS Detection Framework Using Machine Learning Optimized by Bayesian and PSO Techniques
(2026-07-01)
Sathaporn, Posathip
;
Krungseanmuang, Woranidtha
;
Chaowalittawin, Vasutorn
;
Benjangkaprasert, Chawalit
;
Archevapanich, Tuanjai
This paper presents a distributed denial of service (DDoS) detection framework using machine learning techniques enhanced with hyperparameter optimization for network traffic classification and evaluated on the BCCC-cPacket-Cloud-DDoS-2024 dataset. The framework includes data preprocessing with normalization and class imbalance handling via the synthetic minority over-sampling technique. A critical contribution of this study is the rigorous analysis of the trade-off between detection accuracy and model complexity. Unlike arbitrary feature selection methods, we empirically determined the optimal feature set using information gain, identifying that the top 100 features represent the saturation point that balances high accuracy with minimal overhead. Model performance was further improved through hyperparameter optimization using particle swarm optimization and Bayesian algorithms. The extreme gradient boosting (XGBoost) model optimized using Bayesian optimization and the top 100 features achieved the highest performance, with an accuracy of 99.29% and an F1-score of 98.91%. As a result, the proposed framework improves detection performance while reducing model complexity by selecting an optimal feature set to improve model stability and efficiency.
Some of the metrics are blocked by your
consent settings
Physics-Guided CFD–ML Framework for Sustainable Classical Wire Coating with Power-Law Fluids
(2026-07-01)
Nabudda, Kriengkrai
;
Poungthong, Pongthep
;
Ritthong, Wirote
;
Elumalai, P. V.
This study presents an integrated Computational Fluid Dynamics (CFD) and machine learning framework for analyzing and optimizing classical wire coating processes involving non-Newtonian power-law fluids. A two-dimensional axisymmetric CFD model was developed in ANSYS Fluent 2024R1 to investigate the effects of the power-law index (n = 0.3–1.0) on flow, pressure, temperature, and density fields under non-isothermal conditions. A Latin Hypercube Sampling-based Design of Experiments was coupled with surrogate modelling and Sobol sensitivity analysis to evaluate process performance and identify optimal operating conditions. The results showed that velocity distributions were highly dependent on fluid rheology, with shear-thinning fluids producing broader plug-like flow regions and more uniform velocity profiles. In contrast, pressure, temperature, and density fields exhibited limited sensitivity to variations in the power-law index. Optimization indicated that low power-law indices, moderate pressure gradients, and low-to-moderate wire speeds maximize coating thickness while minimizing material loss. Ridge Polynomial Regression achieved excellent predictive accuracy for all response variables (R2 > 0.995). Sensitivity analysis revealed that the initial die gap is the dominant factor governing coating thickness, whereas material loss is influenced by combined effects of die geometry, fluid rheology, and wire speed. The proposed framework provides an efficient tool for process optimization and material conservation in industrial wire coating applications.
Some of the metrics are blocked by your
consent settings
Analysis of Meteorological Influencing Factors and Machine Learning Prediction of Wild Morel Yield in Gannan Prefecture, China
(2026-05-01)
Kong, Dejiang
;
He, Shuling
;
Yongsiri, Ploypailin
Morchella (morels) are rare edible fungi with high economic value but challenging cultivation requirements. Traditional yield prediction methods rely on manual records and statistics, which are time-consuming and yield low accuracy due to morels' long growth cycles, complex environmental factors, and insufficient documentation. This study presents a novel machine learning approach for predicting wild morel yields in Gannan Prefecture, China, using ten years (2013-2023) of monthly meteorological data. Random Forest was applied for feature selection, and Gradient Boosting Machine (GBM) was used to address zero-yield data. The resulting datasets were then employed to develop predictive models with six machine learning algorithms, namely Random Forest, Support Vector Machines (SVM), Long Short-Term Memory (LSTM), Multilayer Perceptron (MLP), Transformer, and Convolutional Neural Network (CNN). Comparative analysis revealed that the GBM-LSTM hybrid model achieved superior performance (MSE: 8047.40, RMSE: 90.43, MAE: 60.90, R2: 0.81). Feature importance analysis identified air humidity (0.298) as the most critical factor affecting morel yields, followed by oxygen concentration, rainfall, light intensity, air quality, CO2 concentration, and average low temperature. Climate trend analysis over the past decade indicates that environmental deterioration, including an increase in temperature (+1.2 °C), a decrease in rainfall (-8.3%), and a reduction in humidity (-6.7%), has been accompanied by a decline in wild morel production, highlighting the vulnerability of this valuable species to climate change. These findings provide scientific guidance for optimizing cultivation strategies and developing climate-adaptive management practices for sustainable morel production.
Some of the metrics are blocked by your
consent settings
Machine Learning-Driven Portfolio Optimization Using Money Flow Index-Based Sentiment Signals
(2026-05-01)
Singsiri, Prapassara
;
Yokrattanasak, Jiraphat
Market indices serve as a benchmark for performance comparison, guide asset allocation decisions, and reflect overall market sentiment and economic conditions, thereby influencing investment strategies by representing a segment of the market. Unquestionably, investor sentiment impacts price movement. In this paper, the objectives were to study the effectiveness of the Money Flow Index (MFI) in enhancing the performance of predictive analysis by capturing market psychology, developing an investment strategy, and analyzing the performance of the method mentioned. This study applies machine learning algorithms with technical indicators and optimizes portfolio allocation based on three notable market indices in Southeast Asia (SEA): SET50 in Thailand, STI in Singapore, and VN30 in Vietnam. Firstly, we combined technical indicators with machine learning—Support Vector Classifier (SVC), Random Forest (RF), and Extreme Gradient Boosting (XGBoost)—by comparing datasets with and without MFI over the period from 2013 to 2023. The results showed that XGBoost with MFI delivered the best predictive performance across three indices. These findings indicate that MFI significantly enhances prediction accuracy, even during volatile market conditions (COVID-19). Additionally, the predictions were integrated into the Markowitz Mean-Variance (MV) model to construct an optimal portfolio, which was then benchmarked against an equal-weight portfolio (1/N). Ultimately, the findings demonstrate that incorporating the machine learning predictions into the MV framework efficiently generates wealth.
Some of the metrics are blocked by your
consent settings
Advancing Masonry Engineering: Effective Prediction of Prism Strength via Machine Learning Techniques
(2026-04-01)
Saingam, Panumas
;
Chatveera, Burachat
;
Nawaz, Adnan
;
Ali, Muhammad Hassan
;
Choudhary, Sandeerah
Masonry buildings have shaped construction history since about 6500 BCE. They offer durability, strength, and cost effectiveness, especially in developing countries. Yet assessing compressive strength during construction remains challenging due to the constituent materials soil, cement, and stone, complicating standardization worldwide. In the present study, an innovative model based on a machine learning algorithm is put forth to predict the compressive strengths of prisms. Some important factors considered as input to the algorithm based on traditional methods are the brick and mortar strengths, prism geometry, mortar bed thickness, and empirically derived height-to-thickness (t) (h/t) ratios. Three different ANN algorithms are coded and trained on the input data, and they are based on the Levenberg–Marquardt algorithm, the resilient backpropagation algorithm, and the conjugate gradient algorithm. The optimal ANN model trained using the conjugate gradient Polak–Ribière algorithm (traincgp) achieves superior performance, with R2 = 0.9881, R2 = 0.9927, RMSE = 0.9914 MPa, MAE = 0.6039 MPa, MAPE = 20.9141%, VAF = 0.9881, and WI = 0.9970. Sensitivity analysis shows the height-to-thickness (h/t) ratio is the dominant influence on compressive strength, consistent with structural mechanics. The primary contributions are the systematically curated, richly parameterized dataset and its use to produce robust, physically interpretable predictions with established ANN methods.
Some of the metrics are blocked by your
consent settings
Predicting Flexural Strength of FRP-Strengthened Waste Aggregate Concrete Beams with Machine Learning: A Step Towards Sustainability
(2026-04-01)
Sangthongtong, Arissaman
;
Chatveera, Burachat
;
Sua-iam, Gritsada
;
Nawaz, Adnan
;
Mehmood, Tahir
Using waste materials in the manufacture of concrete has many environmental advantages. However, it can be difficult to estimate structural performance, especially when beams are reinforced with fiber-reinforced polymers (FRP). In order to provide a data-driven approach to sustainable structural design, this work explores the use of machine learning (ML) approaches to forecast the flexural strength of FRP-strengthened waste aggregate concrete beams. A total number of 92 experimental datasets were used to develop and assess four ML algorithms: Random Forest (RF), Decision Tree (DT), Neural Network (NN), and Extreme Gradient Boosting (XGBoost). Regression plots, Taylor diagrams, statistical measures (R2R^2R2, RMSE, MAE, MSE), and explainable AI (XAI) tools, including SHAP, LIME, and partial dependence plots (PDPs), were used to evaluate the model’s performance. RF outperformed NN in terms of predictive accuracy, while XGBoost exhibited similar performance to RF. The most significant predictors, according to a SHAP analysis, were beam length and fiber length, with the lower followed by steel tensile strength, fiber width, and concrete compressive strength. LIME offered local interpretability for individual predictions, but PDPs demonstrated optimal parameter ranges and a nonlinear feature strength relationship. The findings provide engineers with a strong decision-support tool for designing green infrastructure, since they show that ensemble-based models can accurately represent the intricate, nonlinear dynamics controlling flexural behavior in sustainable FRP-strengthened waste aggregate concrete beams.
Some of the metrics are blocked by your
consent settings
Improving One-Day-Ahead Forecasting of Low-Latitude Amplitude Scintillation Using an Upsampling-Enhanced LSTM
(2026-01-01)
Muangkammuen, Patinya
;
Suthisopapan, Puripong
;
Tongkasem, Napat
;
Supnithi, Pornchai
;
Kruesubthaworn, Anan
The scintillation in radio wave propagation, particularly in regions near the magnetic equator, is found to be introduced by the ionospheric irregularities causing unsatisfactory performance in satellite-based applications. In order to mitigate this effect, we design a long short-term memory (LSTM) model to forecast amplitude scintillation at 1-min resolution. In addition, the upsampling-based feature preprocessing is introduced to improve forecasting performance, especially for short-term severe scintillation events. In terms of R$^{2}$, which is a popular forecast evaluation metric, our proposed model exhibits about 20% improvement over the same LSTM model without upsampling. Furthermore, although existing studies achieve good forecasting accuracy up to 4 h ahead, the proposed model sets a benchmark with one-day-ahead forecasting, but at the cost of longer training time due to upsampling.
Some of the metrics are blocked by your
consent settings
Evaluation of Machine Learning Techniques for Denial-of-Service Attack Detection in Digital Information Exchange
(2026-01-01)
Suwimol, Piyapol
;
Chimmanee, Krishna
;
Pomsathit, Auttapon
Anomaly detection plays a critical role in mitigating cybersecurity threats, particularly Distributed Denial of Service (DDoS) attacks. This study evaluates the performance of tree-based and ensemble learning models, including Decision Tree, Random Forest, and XGBoost, for classifying Snort log data, alongside the application of Isolation Forest for time-series anomaly detection. The experiments were conducted using ICMP-based Ping Flood attacks in a controlled network environment, with data collected from Snort intrusion detection system logs. The classification results indicate that XGBoost achieved the highest performance, with 99.81% accuracy, 99.93% precision, 99.65% recall, and 99.79% F1-score under a 70–30 train-test split. Random Forest and Decision Tree also demonstrated strong performance, while Logistic Regression showed lower effectiveness due to its limitations in modeling nonlinear patterns. For anomaly detection, Isolation Forest was applied to time-series data collected over a 19-day period. The model detected 93 anomaly points, of which 41 overlapped with Wireshark-confirmed events. However, a false positive rate of 41.67% was observed, indicating the need for parameter tuning to balance detection sensitivity and operational efficiency. Overall, the findings demonstrate that ensemble-based learning approaches, particularly XGBoost, are effective for detecting DDoS-related patterns within the experimental setting. However, the results are limited to ICMP-based attack scenarios in a controlled environment. Further validation, including cross-validation, multi-attack evaluation, and deployment-level performance analysis, is required to assess generalizability and practical applicability.
Some of the metrics are blocked by your
consent settings
A Bootstrapping Convolutional Neural Network Technique for Optimizing Automated Detection of Equatorial Plasma Bubbles by Optical All-Sky Imagers
(2025-06-01)
Okoh, Daniel
;
Cesaroni, Claudio
;
Rabiu, Babatunde
;
Shiokawa, Kazuo
;
Otsuka, Yuichi
Equatorial plasma bubbles (EPBs) disrupt satellite-based communication and navigation systems, particularly in equatorial regions. Reliable detection and classification of EPBs from all-sky imager (ASI) images are essential for accurate space weather monitoring and forecasting. This study presents a novel bootstrapping convolutional neural network (CNN) approach to optimize automated EPB detection on ASI images for operational space weather monitoring applications, and overcoming challenges related to image variability and imbalanced data sets. Data used for CNN training were obtained from the optical mesosphere thermosphere imagers ASI installed at the Space Environment Research Laboratory, National Space Research and Development Agency, Abuja during the period from 2015 to 2020. Our method involved training three sub-models, and aggregating their predictions. The CNN trainings were conducted on three sub-datasets of 3,000 images each, categorized as “EPB,” “Noisy/Cloudy” or “No EPB.” Three corresponding sub-models were developed from the CNN trainings. The three sub-model classifications independently gave prediction accuracies of 98.67%, 98.33%, and 95.83% on a reserved test data set of 600 images. Ensemble models further improved the model prediction accuracies to 99.17% and 99.33% for methods based on the mean of sub-model probabilities and the mode of sub-model classifications respectively. Our results indicate that the bootstrapping CNN technique enhanced the EPB detection accuracy, providing a powerful tool for real-time space weather monitoring applications, and implications for improving operational reliability of satellite-based navigation and communication in the equatorial region.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results