KMITL

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

Browse

Search Results

Now showing 1 - 10 of 25

Some of the metrics are blocked by your
consent settings
An Extension of Laor Weight Initialization for Deep Time-Series Forecasting: Evidence from Thai Equity Risk Prediction
(2025-09-01)
Petchpol, Katsamapol
;
Boongasame, Laor
This study presents a gradient-informed proxy initialization framework designed to improve training efficiency and predictive performance in deep learning models for time-series forecasting. The method extends the Laor Initialization approach by introducing backward gradient norm clustering as a selection criterion for input-layer weights, evaluated through a lightweight, architecture-agnostic proxy model. Only the numerical input layer adopts the selected initialization, while internal components retain standard schemes such as Xavier, Kaiming, or Orthogonal, maintaining compatibility and reducing overhead. The framework is evaluated on a real-world financial forecasting task: identifying high-risk equities from the Thai Market Surveillance Measure List, a domain characterized by label imbalance, non-stationarity, and limited data volume. Experiments across five architectures, including Transformer, ConvTran, and MMAGRU-FCN, show that the proposed strategy improves convergence speed and classification accuracy, particularly in deeper and hybrid models. Results in recurrent-based models are competitive but less pronounced. These findings support the method’s practical utility and generalizability for forecasting tasks under real-world constraints.
Some of the metrics are blocked by your
consent settings
Laor Initialization: A New Weight Initialization Method for the Backpropagation of Deep Learning
(2025-07-01)
Boongasame, Laor
;
Muangprathub, Jirapond
;
Thammarak, Karanrat
This paper presents Laor Initialization, an innovative weight initialization technique for deep neural networks that utilizes forward-pass error feedback in conjunction with k-means clustering to optimize the initial weights. In contrast to traditional methods, Laor adopts a data-driven approach that enhances convergence’s stability and efficiency. The method was assessed using various datasets, including a gold price time series, MNIST, and CIFAR-10 across the CNN and LSTM architectures. The results indicate that the Laor Initialization achieved the lowest K-fold cross-validation RMSE (0.00686), surpassing Xavier, He, and Random. Laor demonstrated a high convergence success (final RMSE = 0.00822) and the narrowest interquartile range (IQR), indicating superior stability. Gradient analysis confirmed Laor’s robustness, achieving the lowest coefficients of variation (CV = 0.2230 for MNIST, 0.3448 for CIFAR-10, and 0.5997 for gold price) with zero vanishing layers in the CNNs. Laor achieved a 24% reduction in CPU training time for the Gold price data and the fastest runtime on MNIST (340.69 s), while maintaining efficiency on CIFAR-10 (317.30 s). It performed optimally with a batch size of 32 and a learning rate between 0.001 and 0.01. These findings establish Laor as a robust alternative to conventional methods, suitable for moderately deep architectures. Future research should focus on dynamic variance scaling and adaptive clustering.
Some of the metrics are blocked by your
consent settings
Thai Morning Glory Price Forecasting Using Deep Learning
(2025-01-01)
Waeodi, Kanokwan
;
Boongasame, Laor
;
Thammarak, Karanrat
This study established advanced machine-learning-driven forecasting models to enhance the accuracy of price predictions for Thai morning glory, a widely consumed leafy green vegetable. The models were trained using historical price, weather, and rainfall data using time-series forecasting methods, specifically LSTM and CNN. The findings indicate that stepwise feature selection minimizes prediction errors and improves MSE, RMSE, MAPE, and MAE. Preliminary experiments revealed that the LSTM model with feature selection outperformed the other models, particularly in feature selection. Employing standard hyperparameters of 100 epochs, 32 batches, and five windows, the model demonstrated superior performance with a lower MSE (0.0010), RMSE (0.0274), MAPE (3.7803), and MAE (0.0158) than the CNN model. Statistical hypothesis testing revealed significant variations between the LSTM and CNN models, with feature selection p-values below 0.05. These results indicate that LSTM with feature selection models optimized through refined hyperparameters leads to more accurate Thai morning glory price forecasting, providing valuable insights for stakeholders in their decision-making processes. Additionally, this study can forecast prices for 5, 7, 14, and 21 days in advance based on different Window_len values, addressing various planning needs. The 5- and 7-day forecasts support short-term decision-making, such as scheduling harvest cycles and weekly market planning, whereas the 14-day forecast assists farmers in optimizing planting schedules and logistics. Furthermore, the 21-day forecast is beneficial for medium-term market planning, including negotiating forward contracts and adjusting distribution strategies to maximize profitability.
Some of the metrics are blocked by your
consent settings
Enhancing Predictive Capabilities for Identifying At-Risk Stocks Using Multivariate Time-Series Classification: A Case Study of the Thai Stock Market
(2025-01-01)
Petchpol, Katsamapol
;
Boongasame, Laor
This study proposes a multivariate time-series classification approach using deep learning to predict stocks likely to be flagged by the Market Surveillance Measure List in the Thai stock market. Formulated as a binary classification problem, the model distinguishes At-Risk and Normal stocks based on two primary datasets: End-of-Day stock prices and Market Surveillance Measure List records, incorporating trading volumes and technical indicators. To address data imbalance, concept drift, and long-term dependencies, the framework integrates feature engineering, cost-sensitive learning, and rolling window training. Experimental results show deep learning models significantly outperform traditional baseline methods in capturing financial risk patterns. The study identifies models that effectively balance predictive accuracy with computational efficiency, with performance varying based on forecasting horizons. Despite improvements from specialized techniques, the study identifies challenges in long-term financial risk prediction. These findings support market surveillance, algorithmic trading, and portfolio risk management, with future work exploring explainable AI, adaptive learning, and alternative data sources to enhance interpretability and long-term forecasting.
Some of the metrics are blocked by your
consent settings
Classification of Thai Rice Varieties Using Image Processing and Deep Learning Techniques
(2025-01-01)
Kongmanee, Panpatsorn
;
Puengpradith, Sorapojana
;
Boongasame, Laor
The methods for identifying Thai rice varieties are complex, time-consuming, and require high expertise to achieve accurate results. This research explores different deep learning techniques to efficiently classify the strains of Thai rice that optimize accuracy and speed. The focus rice varieties are Khao Hom-Mali Thai and Thai Hom Pathum Thani 1 fragrant rice; both have similar shapes and characteristics but differ in price, market value, and recognition. The proposed model is based on an instance segmentation model of YOLOv8, which is compared against popular instance segmentation models such as YOLACT, SOLOv2, and Mask R-CNN. Additionally, hyperparameter tuning is performed to ascertain the most optimal values. The evaluation of the model performance reports in the form of mean average precision (mAP), inference time, and model stability. Experimental results indicate that YOLOv8n-seg, with the fewest parameters, achieves the highest accuracy comparable to other YOLOv8-based models with more parameters. The proposed model demonstrates superior accuracy and processing speed performance compared to other state-of-the-art models.
Some of the metrics are blocked by your
consent settings
Enhanced Feature Selection via Hierarchical Concept Modeling
(2024-12-01)
Saelee, Jarunee
;
Wetchapram, Patsita
;
Wanichsombat, Apirat
;
Intarasit, Arthit
;
Muangprathub, Jirapond
The objectives of feature selection include simplifying modeling and making the results more understandable, improving data mining efficiency, and providing clean and understandable data preparation. With big data, it also allows us to reduce computational time, improve prediction performance, and better understand the data in machine learning or pattern recognition applications. In this study, we present a new feature selection approach based on hierarchical concept models using formal concept analysis (FCA) and a decision tree (DT) for selecting a subset of attributes. The presented methods are evaluated based on all learned attributes with 10 datasets from the UCI Machine Learning Repository by using three classification algorithms, namely decision trees, support vector machines (SVM), and artificial neural networks (ANN). The hierarchical concept model is built from a dataset, and it is selected by top-down considering features (attributes) node for each level of structure. Moreover, this study is considered to provide a mathematical feature selection approach with optimization based on a paired-samples t-test. To compare the identified models in order to evaluate feature selection effects, the indicators used were information gain (IG) and chi-squared (CS), while both forward selection (FS) and backward elimination (BS) were tested with the datasets to assess whether the presented model was effective in reducing the number of features used. The results show clearly that the proposed models when using DT or using FCA, needed fewer features than the other methods for similar classification performance.
Some of the metrics are blocked by your
consent settings
Three fakes with deep learning techniques fake news, fake reviewers, and deepfakes: A survey
(2024-10-23)
Boongasame, Laor
The internet has brought convenience to the world. Many people communicate with each other through social media. Truth and lies are among the conveniences that people enjoy. Fake news and fake reviews cause problems for many people, both physically and mentally. For example, mentally, it may cause misunderstandings. Physically, it may lead to incorrect behavior towards the body, such as eating the wrong food or medicine. As a result, this paper presents a survey of research on various false stories that exist on social media. This paper will focus on three distinct topics: 1) fake news; 2) fake reviews; and 3) deepfakes. All three will be surveys of false stories used in deep learning. Some of the three tasks may contain the same information. At the same time, the theory may consist of multiple parts. Next, this chapter will present it in separate parts, hoping that readers will gain an understanding of fake social media and find it useful for future research.
Some of the metrics are blocked by your
consent settings
Design and Implement Deepfake Video Detection Using VGG-16 and Long Short-Term Memory
(2024-01-01)
Boongasame, Laor
;
Boonpluk, Jindaphon
;
Soponmanee, Sunisa
;
Muangprathub, Jirapond
;
Thammarak, Karanrat
This study aims to design and implement deepfake video detection using VGG-16 in combination with long short-term memory (LSTM). In contrast to other studies, this study compares VGG-16, VGG-19, and the newest model, ResNet-101, including LSTM. All the models were tested using Celeb-DF video dataset. The result showed that the VGG-16 model with 15 epochs and 32 batch sizes had the highest performance. The results showed that the VGG-16 model with 15 epochs and 32 batch sizes exhibited the highest performance, with 96.25% accuracy, 93.04% recall, 99.20% specificity, and 99.07% precision. In conclusion, this model can be implemented practically.
Some of the metrics are blocked by your
consent settings
A Comparative Study of Sentiment Analysis Methods for Detecting Fake Reviews in E-Commerce
(2023-06-01)
Puttarattanamanee, Maneerat
;
Boongasame, Laor
;
Thammarak, Karanrat
The popularity of the e-commerce system has increased, especially under the COVID scenario. Consumer product reviews from the past have had a significant impact on influencing consumers' purchasing decisions. Fake reviews—those written by humans and computers that engage in dishonest behavior—are consequently generated to increase product sales. The fake reviews hurt consumers and are dishonest. The goal of this research is to examine and evaluate the performance of various methods for identifying fake reviews. The well-known and widely-used Amazon Review Data (2018) dataset was used for this research. The first 10 product categories on Amazon.com with favorable feedback will be provided in the data section. After that, perform fundamental data preparation procedures such as special character trimming, bag of words, TF-IDF, etc. The models are trained to create a dataset for detecting fake reviews. This research compares the performance of four different models: GPT-2, NBSVM, BiLSTM, and RoBERTa. The hyperparameters of the models are also tuned to find the optimal values. The research concludes that the RoBERTa model performs the best overall, with an accuracy of 97%. GPT-2 has an overall accuracy of 82%, NBSVM has an overall accuracy of 95%, and BiLSTM has an overall accuracy of 92%. The research also calculates the Area Under the Curve (AUC) for each model and finds that RoBERTa has an AUC of 0.9976, NBSVM has an AUC of 0.9888, BiLSTM has an AUC of 0.9753, and GPT-2 has an AUC of 0.9226. It can be observed that the RoBERTa model has the highest AUC value, which is close to 1. Therefore, it can be concluded that this model provides the most accurate prediction for detecting fake reviews, which is the main focus of this research.
Some of the metrics are blocked by your
consent settings
Gold-Price Forecasting Method Using Long Short-Term Memory and the Association Rule
(2023-01-01)
Boongasame, Laor
;
Viriyaphol, Piboonlit
;
Tassanavipas, Kriangkrai
;
Temdee, Punnarumol
Since gold prices influence international economic and monetary systems, numerous studies have been conducted to forecast gold prices. Nonetheless, studies employing the linear relationship method usually fail to explain the change in the pattern of the gold price. This study introduces a new paradigm that incorporates association rules and long short-term memory (LSTM) as a nonlinear-based method. For simulation, the proposed method was analyzed with data from Yahoo Finance from January 2010 to December 2020. The association rule was used to choose features relevant to the gold spot (GS) in the US Dollar Index (DXY). The LSTM forecast the gold price with a range of hyperparameter settings. The simulation results showed that the proposed method—the LSTM with GS and DXY, or LSTM-GS-DXY—resulted in low mean absolute percentage error (MAPE) metrics. In addition, the proposed LSTM-GS-DXY system outperformed the simple moving average (SMA), weight moving average (WMA), exponential moving average (EMA), and auto-regressive integrated moving average (ARIMA).

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results