KMITL

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

Browse

Search Results

Now showing 1 - 10 of 16

Some of the metrics are blocked by your
consent settings
A Comparative Study of NLP-Based Models: Popularity Trends in Pop Mart Comments
(2026-05-04)
Charoenporn, Pattama
;
Poonsawat, Kititinun
;
Banjongkan, Anupong
In today's digital era, where consumer sentiment significantly influences brand perception, understanding customer emotions has become a strategic priority for companies like POP MART. Among its iconic collectibles, Labubu stands out as a symbol of the brand's global appeal, gaining tremendous popularity through the viral influence of Lisa from BLACKPINK, who showcased her Labubu collection on social media. This moment sparked a surge in interest, especially in Thailand, where Labubu achieved over 365,000 mentions in a single month, driven by its appeal to young, trend-conscious consumers. This study, Sentiment Analysis of LABUBU Popularity Trends: A Study of POPMART Comments Using NLP Techniques, investigates customer perceptions and emotions toward Labubu by employing advanced transformer-based NLP models - WangchanBERTa, RoBERTa, and XLM-R. By analyzing Thai-specific and multilingual datasets collected from social media platforms, this research evaluates the models' performance in sentiment classification. The analysis highlights WangchanBERTa's superior performance on Thai-specific datasets, achieving the highest F1-score of 87.51%. XLM-R demonstrated robust multilingual capabilities with an F1-score of 84.65%, while RoBERTa excelled in English sentiment analysis with an F1-score of 82.83%. These findings emphasize the importance of leveraging both language-specific and multilingual models to address cultural and linguistic nuances effectively. Sentiment patterns reveal positive comments highlighting emotional connections to Labubu, neutral sentiments about logistical inquiries, and negative feedback on technical and availability issues. These insights guide strategies to enhance marketing, boost customer engagement, and address operational challenges. This study advances NLP applications in sentiment analysis, offering POP MART data-driven insights to refine global and regional strategies while underscoring the value of multilingual transformer models in analyzing diverse consumer feedback.
Some of the metrics are blocked by your
consent settings
Performance Evaluation of Imputation Techniques for Telecommunications Customer Clustering
(2026-01-01)
Sukthong, Patthama
;
Charoenporn, Pattama
Missing data significantly degrades machine learning model performance in telecommunications customer analytics, leading to unreliable customer segmentation and suboptimal business decision-making. This research systematically compares seven imputation techniques across three missing mechanisms (MCAR, MAR, MNAR) and four missing rates (5%, 10%, 20%, 30%) using the Telco Customer Churn Dataset (7,043 records). Methods evaluated include traditional approaches (mean/mode, forward fill, regression), machine learning techniques (KNN, Random Forest, MICE), and deep learning (Autoencoder). We assessed model performance using normalized MAE and RMSE, and evaluated downstream effects through clustering algorithms. Results demonstrate Random Forest imputation's superior performance with MAE of 0.1568 and RMSE of 0.2123, achieving 53.7% lower error rates compared to mean/mode imputation. Statistical analysis confirmed significant performance differences (Friedman test: χ<sup>2</sup> = 55.85, p < 0.001). Interestingly, clustering performance did not directly correlate with imputation accuracy; the Autoencoder achieved the highest silhouette score (0.1510) despite moderate reconstruction accuracy. Machine learning approaches maintained robust performance across all missing data mechanisms, whereas traditional methods degraded under MNAR conditions. These findings provide evidence-based guidelines for selecting appropriate imputation techniques in telecommunications analytics, enabling improved customer segmentation and business outcomes.
Some of the metrics are blocked by your
consent settings
Early Diagnosis of Knee Osteoarthritis With a Natural Language Processing–Driven Approach Based on Clinician Notes: Development and Validation Study
(2025-01-01)
Thanyakunsajja, Narathip
;
Jitkajornwanich, Kulsawasd
;
Xu, Shan
;
Shin, Donghee
;
Charoenporn, Pattama
Background: Knee osteoarthritis (OA) is a common form of knee arthritis that can cause significant disability and affect a patient’s quality of life. Although this disease is chronic and irreversible, the patient’s condition can be improved and the progression of the disease can be prevented if the disease is diagnosed early and the patient receives appropriate treatment immediately. Therefore, the prediction of knee OA is considered one of the essential steps to effectively diagnose and prevent further severe OA conditions. Knee OA is commonly diagnosed by medical experts or physicians, and the diagnosis of OA is mainly based on patients’ laboratory results and medical images, including x-ray and magnetic resonance images. However, diagnosis through such data is often time-consuming. Moreover, the diagnosis results can vary among physicians depending on their expertise. Previous studies mostly focused on using approaches, such as those involving artificial intelligence, to automatically detect knee OA through such data. However, these studies did not incorporate clinicians’ or doctors’ notes (text data) into the analysis, although these data involving reported symptoms and behaviors are already available and easier to collect and access than laboratory data and image data. Objective: We propose a novel natural language processing–driven approach based on clinicians’ or doctors’ notes of patient-reported symptoms (text data only) for diagnosing knee OA. Methods: The textual information from clinicians’ or doctors’ notes was first preprocessed using text analysis algorithms with respect to natural language processing. We then incorporated deep learning models, including convolutional neural networks, bidirectional long short-term memory (BiLSTM), and gated recurrent units. Lastly, a disease-specific standard questionnaire called WOMAC (Western Ontario and McMaster Universities Arthritis Index) was taken into account to improve the overall performance of the models. Results: Our experiment included 5849 records (OA: 3455; non-OA: 2394). Before applying our WOMAC-based processing approach, the best-performing model was BiLSTM (area under the curve, 0.85; accuracy, 0.87; precision, 0.85; sensitivity, 0.95; specificity, 0.76; F<inf>1</inf>-score, 0.90), and there was an improvement in the results with BiLSTM after applying our approach (area under the curve, 0.91; accuracy, 0.91; precision, 0.91; sensitivity, 0.94; specificity, 0.87; F<inf>1</inf>-score, 0.93). Conclusions: Our proposed method for predicting the occurrence of knee OA showed better performance than other conventional methods that use image data and statistical laboratory data. The findings indicate the feasibility of using text data (symptom descriptions reported by patients and recorded by doctors) to predict knee OA. Medical notes of symptom reports can be considered a valuable data source for predicting whether a particular knee is likely to experience OA progression.
Some of the metrics are blocked by your
consent settings
Visualizing Political Communication Trends across Generations on X (Twitter): Insights Through Topic Modeling and Word Clouds
(2024-01-01)
Udomwisanpat, Prinwat
;
Jitkajornwanich, Kulsawasd
;
Kraishan, Obada
;
Srestasatheirn, Panu
;
Lawawirojwong, Siam
This study examines the interests and significance of words on Twitter (or X) across different generational groups: Baby Boomers, Generation X, Generation Y, and Generation Z. Using Topic Modeling with Latent Dirichlet Allocation (LDA), the research explores relationships and word importance within each group. As the results of topic modeling are not always easy to interpret, we used word cloud visualization to help make sense of the results for each generation. The findings reveal distinct patterns: Baby Boomers frequently mention print media, news websites, and prominent Thai political figures; Generation X emphasizes individuals and local political issues in Bangkok; Generation Y discusses political and social events; and Generation Z uniquely questions political and social activities. This research methodology is applicable across languages and tasks, offering insights into generational behaviors and interests.
Some of the metrics are blocked by your
consent settings
Knowledge Discovery and Dataset for the Improvement of Digital Literacy Skills in Undergraduate Students
(2023-07-01)
Nilaphruek, Pongpon
;
Charoenporn, Pattama
For over two decades, scholars and practitioners have emphasized the importance of digital literacy, yet the existing datasets are insufficient for establishing learning analytics in Thailand. Learning analytics focuses on gathering and analyzing student data to optimize learning tools and activities to improve students’ learning experiences. The main problem is that the ICT skill levels of the youth are rather low in Thailand. To facilitate research in this field, this study has compiled a dataset containing information from the IC3 digital literacy certification delivered at the Rajamangala University of Technology Thanyaburi (RMUTT) in Thailand between 2016 and 2023. This dataset is unique since it includes demographic and academic records about undergraduate students. The dataset was collected and underwent a preparation process, including data cleansing, anonymization, and release. This data enables the examination of student learning outcomes, represented by a dataset containing information about 45,603 records with students’ certification assessment scores. This compiled dataset provides a rich resource for researchers studying digital literacy and learning analytics. It offers researchers the opportunity to gain valuable insights, inform evidence-based educational practices, and contribute to the ongoing efforts to improve digital literacy education in Thailand and beyond. Dataset: https://dx.doi.org/10.21227/370s-1s37 Dataset License: CC-BY 4.0
Some of the metrics are blocked by your
consent settings
The Development of the Logistics Transportation Management Prototype System: A Case Study of the Economic Development along the East-West Economic Corridorin Thailand
(2022-10-01)
Charoenporn, Pattama
The objective of this research is to design a logistic transportation management system prototype that can manage operations through a one-stop service. The development of the prototype will be utilizing a Saving Algorithm and Google Map to help analyze routes and determine the lowest cost. The design of the various functions will be based on samples of the transportation data of the East-West Economic Corridor. The research’s data collection will be based on data obtained from 400 transport and logistical operator samples, professionals who possess knowledge related to logistic transportation management, or academics who are well- versed in logistic operations. The collection of data was conducted using questionnaires, which were used to conduct interviews before the design of the prototype, and thereafter, to evaluate the completed logistic system design. As a result of the data collection, the researcher was then able to use the information obtained to design the interface of the logistic transportation management prototype system, which was able to comprehensively define the various functions of the prototype system. The results from testing the system based on the ISO 9241-151 principles showed that the detailed layout of the screens was rated to be Very Good by 4 persons, while 2 persons rated it as Fair. The reason that it was rated as Fair was due to the details of the proposed content was not very comprehensive as expected. As for the aspect of searching for information, it was rated as Very Good by 2 points, The display of the contents and details of the system function, such as fonts or colors, were rated to be Very Good by 3 points, and Good by 2 points. Experts commented that the system should be able to provide more details in explaining the functions of the different menus on each page or adding a navigator for other topics, such as the system processing speed. The system’s processing function was rated as Good by 3 points and Fair by 2 points.Finally, the results show that, the prototype has presented guidelines for applying the concepts of both algorithms to work collaboratively with the logistics system.It canhelp transporter to find the route of travel as well.
Some of the metrics are blocked by your
consent settings
Automatical IVUS Image Segmentation and Detection Using BLOB Analysis
(2022-01-01)
Onhirun, Nippitch
;
Charoenporn, Pattama
In this research, experiments were conducted on the problem of image analysis in atherosclerotic disease. The study was conducted from intravascular ultrasound images that have been taken inside the blood vessels from patients with arterial wall hardening or stenosis. These images are used by experts or specialists to analyze the problem of the disease. There may be some interference from shadows or elements that occurred while taking the image that may cause the incomplete image. We propose an applied method by using computer techniques that help to analyze the components within the image to find the area of interest by enhancement image. Then, the pixel levels are analyzed and classified to divide the background from the image. Next, the morphological operation is used to readjust image properties, and blob analysis is used to identify the region of interesting pixel values. In the final process, Connected Component Analysis (CCA) was used to detect media-adventitia and lumen boundary area The efficacy of segment results was measured by comparing with the expert to measure performance as (media-adventitia, lumen) the Jaccard Index = (0.9570, 0.96695), Hausdorff Distance = (0.5259, 0.6304), Percentage Area Distance = (0.0382, 0.03395) IVUS dataset was used from Simone Balocco, Dept. Matemàtica Aplicada i Anàlisi, Universitat de Barcelona, Barcelona, Spain.
Some of the metrics are blocked by your
consent settings
Software testing system development based on ISO 29119
(2021-05-01)
Raksawat, Chadatarn
;
Charoenporn, Pattama
The software testing process is an important method in various fields. Every task in software process must be tested before delivery to the customer. So, in the software field, a testing process is necessary for create application. Today, many researchers find a testing methodology from a software testing standard that will ensure many people around the world. This paper chooses the ISO 29119 standard to create a prototype for the testing process. It is suitable for small business and guides developer to generate their software. The result of paper is shown step to test software and creates document to compare results between an actual result from user and an expected result from ISO 29119 standard. Finally, the system is tested with black box methods and evaluated by the specialist that test user satisfaction survey. The results appear a satisfaction average value at 78.4%.
Some of the metrics are blocked by your
consent settings
The Development of the Wireframes Design for Usability under the ISO 9241-151 Standard
(2020-06-05)
Charoenporn, Pattama
This research aims to design a wireframe according to the usability test and the ISO 9241-151 standard. The procedure starts with collecting data from 400 website users with questions under the ISO 9241-151 standard and collecting data from interviews with Design Thinking technique from 20 web designers with evaluation interviews based on the System Usability Scale from 3 websites chosen as the sample space for data collection; namely sanook.com, kapook.com and dex-d.com. After analyzing the data from the sample group, a wireframe and site map is then built. Then, 5 experts will test the wireframe design based on the hypothesis created by the research. The results show that analysis of sample sites usage has a significant statistical relationship with wireframe design at level 0.05. With the System Usability Scale, wireframes and site maps have been designed to be assessed by 5 design specialists. Results appear that Header and Content design have been rated as Good by 3 specialists and Satisfactory by 2. Navigation and Information search was rated as Very Good by 2 and Good by 3. The content display has been rated as Very Good by 2 and Good by 3. Other topics have been rated as Very Good by 3 and Good by 2. Compiling those results conclude that this wireframe design is efficiently operational under the ISO 9241-151 standard, allowing design enthusiasts to collect information from this research to guide their future site designs. In the future, there should be some additional of other types of website design to cover more of the website designers' needs.
Some of the metrics are blocked by your
consent settings
UX-Based Design of A Mobile Application for Thai Seniors
(2020-06-05)
Aranyanak, Inthraporn
;
Charoenporn, Pattama
Mobile technology has become a vital part of our lives. There is a large number of studies to develop mobile applications. Nevertheless, some of them were not designed for the elderly to use easily. By 2021, Thailand will become a full-fledged ageing society. Workers aged more than 45 will be at greater risk of losing their job because of economic recession or the growth of AI technology. This paper demonstrates the usability testing of mobile job applications comparing three popular job apps used in Thailand. An analytical tool for prototypes and the System Usability Scale (SUS) were used to collect and analyse data. The main usability metrics in this study are effectiveness, efficiency, and satisfaction. The results show that using ambiguous labels, icons with multiple meanings, and using a "hamburger" icon can cause Thai senior users to become disoriented and affect their performances.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results