检索结果-内蒙古大学图书馆

6th International Conference on Mathematics and Statistics, ICoMS 2023

作者： Na Bangchang, Kannat Wongsai, Sangdao Simmachan, Teerawat Department of Mathematics and Statistics Faculty of Science and Technology Thammasat University Pathum Thani Thailand Department of Mathematics and Statistics Thammasat University Research Unit in Data Learning Faculty of Science and Technology Thammasat University Pathum Thani Thailand

ISBN: (纸本)9798400700187

The insurance industry is a fast-growing industry and handles substantial amounts of data. Fraudulent claims are the main problem in the industry. Auto insurance fraud is one of the most prominent types of insurance fraud. Numerous fraudulent claims affect not only the insurance company but also the sincere policy holders because of the increasing in premium amounts. Therefore, detection of insurance fraud is a challenging problem. Traditional approaches are hard to handle and inefficient. data mining has recently offered significant contributions to insurance analysis. To overcome this, data mining techniques are used to predict fraudulent claims. This work would help in a screening process to investigate claims, thus minimizing human resources and monetary losses. Three sets of features are obtained by logistic regression models: one with forward selection, one with backward elimination, and one without variable selection. Three algorithms including Naïve Bayes, random forest and adaptive boosting are employed as classifiers. K-fold cross validation is used to evaluate the algorithm performance. The results suggest that the smaller number of features, the better performance. The random forest performs the best with highest accuracy (85.28%), sensitivity (93.85%), and precision (97.91%) whereas the adaptive boosting provides the highest specificity (70.41%) and F-score (89.86%). © 2023 ACM.

关键词： Adaptive Boosting Naïve Bayes Random Forest Variable Selection

来源：评论

学校读者我要写书评

暂无评论

Improving AI-Based Skin Disease Classification with StyleGAN3 for Minority Skin Tone Generation 16

Improving AI-Based Skin Disease Classification with StyleGAN...

引用

16th Biomedical Engineering International Conference, BMEiCON 2024

作者： Tanantong, Tanatorn La-Or-On, Pangon Srijiranon, Krittakom Thammasat University Research Thammasat University Unit in Data Innovation and Artificial Intelligence Faculty of Science and Technology Department of Computer Science Pathum Thani Thailand

ISBN: (纸本)9798331505431

Skin diseases are prevalent in Thailand due to the hot temperatures, humid climate, and increasing diversity of skin tones, which complicates diagnosis. Artificial intelligence (AI) has improved skin disease detection through deep learning models, aiding faster diagnoses and reducing the workload on dermatologists. However, AI models often struggle with underrepresentation of darker skin tones in training datasets, affecting performance. To address this imbalance, data augmentation techniques like generative models are recommended. This research enhances skin disease classification using StyleGAN3 for data augmentation to balance minority skin tone classes. The Fitzpatrick17k dataset focusing on Acne, Psoriasis, and Vitiligo is used. The Fitzpatrick Skin Type Scale is reduced from six to three levels due to image availability. VGG16 is applied to classify skin tone classes. Two additional datasets, the original and traditionally augmented datasets, are used for comparison against the StyleGAN3-augmented dataset. Results show that the StyleGAN3-augmented dataset achieves the best performance with F1-scores of 0.7193, 0.7446, and 0.7543 and accuracy scores of 0.7200, 0.7368, and 0.7419 for Acne, Psoriasis, and Vitiligo respectively. The obtained results demonstrate that utilizing the StyleGAN3-augmented dataset for skin disease classification in minority skin tone classes yields the highest performance, with F1-scores increasing by 19.25%, 19.77%, and 25.94 % for Acne, Psoriasis, and Vitiligo, respectively. These findings highlight the effectiveness of StyleGAN3 in improving model performance for underrepresented skin tones in skin disease classification. © 2024 IEEE.

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

An IoT-based Bearded Dragon Enclosure Integrating YOLO Object Detection for Monitoring Food and Cleanliness 12

An IoT-based Bearded Dragon Enclosure Integrating YOLO Objec...

引用

12th IEEE Global Conference on Consumer Electronics, GCCE 2023

作者： Varisthanist, Nanmanat Tardtong, Thanapat Jadee, Kittitach Srijiranon, Krittakom Thammasat University Faculty of Science and Technology Thammasat University Research Unit in Data Innovation and Artificial Intelligence Department of Computer Science Pathumthani Thailand

ISBN: (纸本)9798350340181

Among exotic reptiles, bearded dragons (Pogona) are popular pets. As ectotherms, bearded dragons require special food and controlled environments, including temperature, humidity, ultraviolet light, and cleanliness. This research applied the Internet of Things (IoT) to control environments within the pet enclosure. In addition, the You Only Look Once (YOLO) deep learning algorithm applied to convolutional neural networks was used to create object detection to count crickets eaten by a dragon and detect feces inside the enclosure. To find the appropriate YOLO structure, three varied experiments were proposed, including different versions YOLO and sizes as well as hyperparameter tuning. Results were that YOLYv5s with 16 batches and 0.01 learning rate was the appropriate structure with an average precision of 0.8246 and 0.6051 for detecting crickets and feces, respectively, at a mean average precision of 0.7149. The proposed enclosure with object detection may improve efficiency and reduce redundancy in breeder tasks. © 2023 IEEE.

关键词： Enclosures

来源：评论

学校读者我要写书评

暂无评论

Thai Conversational Chatbot Classification Using BiLSTM and data Augmentation 1st

Thai Conversational Chatbot Classification Using BiLSTM and ...

引用

1st International Conference on data Science and Artificial Intelligence, DSAI 2023

作者： Lhasiw, Nunthawat Tanantong, Tanatorn Sanglerdsinlapachai, Nuttapong Thammasat Research Unit in Data Innovation and Artificial Intelligence Department of Computer Science Faculty of Science and Technology Thammasat University Pathum Thani Thailand Strategic Analytics Networks with Machine Learning and AI Research Team National Electronics and Computer Technology Center Pathum Thani Thailand

ISBN: (纸本)9789819979684

Chatbot platforms, e.g., Facebook and Line, have revolutionized human interaction in the digital age. In order to develop an automatic chatbot classification, there are several challenges especially for Thai chat messages. Conversational messages are usually short and ambiguous. Therefore, it is difficult to find a dataset for constructing an effective classification model. To address the limited size of the dataset, data augmentation techniques can be possibly applied. data augmentation involves generating synthetic messages by applying various transformations to existing data samples while preserving their original meaning. In this study, the size and diversity of the dataset is increased by two methods, i.e., text augmentation using word2vec from Thai2Fit and English-Thai machine translation models proposed by VISTEC. Based on the augmented messages, a Deep learning technique, BiLSTM, is used to construct a chatbot classification model. The experimental obtained results demonstrate that data augmentation can help to increase the classification performance. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Improving AI-Based Skin Disease Classification with StyleGAN3 for Minority Skin Tone Generation

Improving AI-Based Skin Disease Classification with StyleGAN...

引用

Biomedical Engineering International Conference (BMEiCON)

作者： Tanatorn Tanantong Pangon La-or-on Krittakom Srijiranon Department of Computer Science Unit in Data Innovation and Artificial Intelligence Faculty of Science and Technology Thammasat University Research Thammasat University Pathum Thani Thailand

ISBN: (数字)9798331505431

ISBN: (纸本)9798331505448

关键词： Training Accuracy Psoriasis Focusing data augmentation Skin data models Artificial intelligence Diseases Meteorology

来源：评论

学校读者我要写书评

暂无评论

An IoT-based Bearded Dragon Enclosure Integrating YOLO Object Detection for Monitoring Food and Cleanliness

An IoT-based Bearded Dragon Enclosure Integrating YOLO Objec...

引用

IEEE Global Conference on Consumer Electronics (GCCE)

作者： Nanmanat Varisthanist Thanapat Tardtong Kittitach Jadee Krittakom Srijiranon Thammasat University Research Unit in Data Innovation and Artificial Intelligence Department of Computer Science Faculty of Science and Technology Thammasat University Pathumthani Thailand

关键词：

来源：评论

学校读者我要写书评

暂无评论

Proposing R&D Bridge Manager Competency Development Framework

Proposing R&D Bridge Manager Competency Development Framewor...

引用

2022 IEEE Technology and Engineering Management Society Conference - Asia Pacific, TEMSCON-ASPAC 2022

作者： Chalarak, Nawarerk Uchihira, Naoshi Faculty of Science and Technology Thammasat University Thammasat Research Unit in Data Innovation and Artificial Intelligence Department of Computer Science Pathum Thani Thailand School of Knowledge Science Japan Advanced Institute of Science and Technology Ishikawa Japan

ISBN: (数字)9781665427692

ISBN: (纸本)9781665427692

With the increasing demand for global R&D collaboration to utilize knowledge resources around the world, a competency development framework (CDF) for managers who facilitate research collaboration between R&D teams in different countries is necessary for the managers to perform their jobs effectively and to improve the efficiency of utilizing global knowledge resources. This paper proposes a CDF for R&D bridge managers (BMs) who bridge R&D organizations in different countries. A CDF assists with designing professional development programs. The proposed framework is based on relevant literature and empirical studies. It defines competencies that would most likely help BMs enhance research collaboration in global R&D projects. This is similar to how the project manager competency development framework impacts the performance of a project manager. A workshop method is proposed as a competency development method for BMs. This paper presents the first R&D bridge manager competency development framework (BMCDF) for BMs who facilitate research collaboration in global R&D projects. The managers can use the BMCDF for their professional development, especially in the context of global R&D projects. organizations may use this framework as a guide for developing competency-based training programs for their R&D personnel. © 2022 IEEE.

关键词： Managers

来源：评论

学校读者我要写书评

暂无评论

Elderly Job Opportunity Analytic and Prediction System for Thai Aging Society 26

Elderly Job Opportunity Analytic and Prediction System for T...

引用

26th International Computer Science and Engineering Conference, ICSEC 2022

作者： Luangwiriya, Taweewat Siriphorn, Akkradate Saithanu, Somying Fakthong, Tiraphap Pensri, Praneet Leckcivilize, Attakrit Mongkhonittivech, Nongyao Kongkachandra, Rachada College of Interdisciplinary Studies Thammasat University Data Science and Innovation Thailand Chulalongkorn University Faculty of Allied Health Sciences Bangkok Thailand Think Change Learning Social Enterprise Co. Ltd. Nonthaburi Thailand Thammasat University Faculty of Economics Thailand Chulalongkorn University Faculty of Allied Health Sciences Department of Physical Therapy Thailand University of Aberdeen Health Economics Research Unit Aberdeen United Kingdom Mae Fah Luang University Adult Nursing Department School of Nursing Chiang Rai Thailand

ISBN: (纸本)9781665491983

Thailand is now a complete aging society, which impact the society and economy. A measure to provide stable income for elders is to support health promotion together with upskilling and reskilling of working. The objectives of this research are to crate and develop a system to recommend suitable job for elders and analyze the factors that motivate elders to work in a local area. The system provides services of record data, analyze data, processing data, visual analytics, and even predict the Elders' working ability for up to 76.6% accuracy in overall. The top factors were analyzed and found to indicate a decision and motivation behind the current status of elder working situation. © 2022 IEEE.

关键词： data handling

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Effects of Different Urban and Rural Characteristics on Surface Urban Heat Island Intensity Using the Spatial Autoregressive Model

SSRN

引用

SSRN 2023年

作者： Wongsi, Noppachai Wongsai, Sangdao Wanishsakpong, Wandee Suwanprasit, Chanida Department of Mathematics and Statistics Faculty of Science and Technology Thammasat University Pathumthani12121 Thailand Thammasat University Research Unit in Data Learning Thammasat University Pathumthani12121 Thailand Department of Statistics Faculty of Science Kasetsart University Bangkok10900 Thailand Department of Geography Faculty of Social Sciences Chiang Mai University Chiang Mai50200 Thailand College of Arts Media and Technology Chiang Mai University Chiang Mai50200 Thailand

This study introduces an approach for quantifying the SUHII using the generalized mixed spatial autoregressive (SAR) model, in the Eastern Economic Corridor of Thailand. A comparative analysis was performed using the mean differences and simple linear regression (SLR), SLR with kernel density estimation (KDE), and spatial autoregressive regression (SAR) models. A year of MODIS LST was the dependent variable. The independent variables were the ISA percentage (%ISA), derived from the Sentinel-2 image using the RF classifier, and the geographical locations. Our findings indicated that using the mean differences on pixels without water surface and altitude led to the underestimated SUHII. By fitting the SLR with KDE on the smoothed %ISA, the LST was overfitted. The SLR applied to the original dataset mitigated the overestimation issue, but could not address the spatial dependency. The SAR model revealed a significant reduction in the spatial dependency and enhanced the model accuracy without overfitting. The 95% confidence intervals for LST from the SAR were found to be twice as narrow as those from other models. To evaluate the spatial assumption, we suggested making a map of how the spatial residuals are spread out, showing if there are clusters or dispersed. © 2023, The Authors. All rights reserved.

关键词： Atmospheric temperature

来源：评论

学校读者我要写书评

暂无评论

Stroke Home Rehabilitation Approach Using Mobile Application Based on PostNet Machine learning Model 23

Stroke Home Rehabilitation Approach Using Mobile Application...

引用

7th International Conference on Medical and Health Informatics, ICMHI 2023

作者： Das, Utpal Chandra Le, Ngoc Thien Benjapolakul, Watit Vitoonpong, Timporn Pluempitiwiriyawej, Charnchai Center of Excellence in Artificial Intelligence Machine Learning and Smart Grid Technology Department of Electrical Engineering Chulalongkorn University Bangkok10330 Thailand Department of Rehabilitation Medicine Faculty of Medicine Chulalongkorn University Bangkok10330 Thailand Multimedia Data Analytics and Processing Research Unit Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok10330 Thailand

ISBN: (纸本)9798400700712

Stroke is a significant cause of mortality and disability globally, with its occurrence in the human brain and motor function being linked to various parts of the human body. Stroke victims often experience disabilities or mobility problems in affected body parts, either on one or both sides of the body. Physiotherapy exercises are the primary treatment and medication for stroke patients, necessitating daily monitoring by a physiotherapist. However, this approach is expensive, and rehabilitation centers and physicians are scarce. Numerous research studies have been conducted to address this issue from various perspectives. This study proposes a Convolutional Neural Network (CNN) based pose net machine learning (ML) model for stroke home rehabilitation using pose detection and classification with a skeleton base model and human pose estimation drawing. We trained our ML model using different human pose images for this model. The tested accuracy of our CNN model is 100% for our exercise pose image test case. Later we build and mobile application for remote rehabilitation. Tested the application with ten different subjects, achieving a 98% for elbow extension, and in the case of elbow flexion and Normal position of balancing both sides, all the persons achieved a 100 % accuracy rate under a laboratory environment. © 2023 ACM.

关键词： Patient treatment

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：