检索结果-内蒙古大学图书馆

International Joint Symposium on Artificial Intelligence and Natural language Processing (iSAI-NLP)

作者： Phonlaphat Na Pompech Wasan Na Chai Peerachet Porkaew Nattapol Kritsuthikul Princess Chulabhorn Science High School Phetchaburi Phetchaburi Thailand Language and Semantic Technology Laboratory (LST) National Electronics and Computer Technology Center (NECTEC) Thailand

ISBN: (数字)9798331509910

ISBN: (纸本)9798331509927

This study explores the effectiveness of genre-specific machine translation (MT) in translating English movie subtitles into Thai, with a focus on the romance and horror genres. Standard MT models often fall short in capturing the unique nuances and context-specific language of different genres, potentially affecting translation quality. To address this, we collected parallel subtitle data for romance and horror films from the OpenSubtitles 2018 dataset. We systematically compared three MT approaches on the two genres. By fine-tuning the pretrained model on genres-specific dataset, we obtain +1.7 and +1.1 BLEU improvements on Romance and Horror, respectively. The results demonstrate that genre-specific models yield higher accuracy when applied to genre-matched content, underscoring the value of tailored MT models for enhancing subtitle translations.

关键词： Translation Accuracy Films Motion pictures Machine translation Artificial intelligence Standards Context modeling

来源：评论

学校读者我要写书评

暂无评论

Improving CNN-Based Blood Pressure Estimation from PPG Data Through Multi-Window Prediction Averaging and Interval Classification

Improving CNN-Based Blood Pressure Estimation from PPG Data ...

引用

International Joint Symposium on Artificial Intelligence and Natural language Processing (iSAI-NLP)

作者： Akira Sakashita Worasawate Apichart Intarapanich Anongnat Khamkla Center for Educational Research and Development Kasetsart University Laboratory School Bangkok Thailand National Electronic and Computer Technology Center National Science and Technology Development Agency Pathum Thani Thailand

ISBN: (数字)9798331509910

ISBN: (纸本)9798331509927

Accurate and continuous blood pressure (BP) monitoring using non-invasive methods is critical for various clinical applications. However, traditional approaches can be uncomfortable and resource-intensive, limiting their usability in continuous monitoring settings. This study utilizes a machine learning model to estimate systolic blood pressure (SBP) and diastolic blood pressure (DBP) from photoplethysmogram (PPG) data derived from the MIMIC-III database. The previously established method for predicting BP from seven-second PPG windows is applied, and an extended approach is proposed using overlapping six-second windows for multi-window prediction averaging. Various prediction strategies are systematically compared, including single predictions from both seven-second and six-second data, as well as mean and median multiple predictions derived from six-second data, along with a pre-trained model. Additionally, a BP interval classification model with mean regression is introduced and evaluated separately, demonstrating a further reduction in mean absolute error (MAE) and suggesting that finer BP intervals could enhance prediction accuracy. Experimental results indicate that multi-window averaging and interval classification improve accuracy, highlighting the potential of these methods for advancing non-invasive continuous BP monitoring.

关键词： Training Accuracy Estimation Predictive models Blood pressure Data models Reliability Biomedical monitoring Usability Monitoring

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph Enhanced Community Consensus: a Scenario-based Knowledge Construction on Buddha Images 21

Knowledge Graph Enhanced Community Consensus: a Scenario-bas...

引用

Proceedings of the 10th International Joint Conference on Knowledge Graphs

作者： Akkharawoot Takhom Tharathon Utasri Dhanon Leenoi Pitchaya Soomjinda Prachya Boonkwan Thepchai Supnithi Language and Semantic Technology Laboratory National Electronics and Computer Technology Center Thailand Faculty of Science and Technology Rajamangala University of Technology Thailand Faculty of Fine Arts Chiang Mai University Thailand

来源：评论

学校读者我要写书评

暂无评论

Fine-tuning Transformer-based MT Using Syntactic Guides

Fine-tuning Transformer-based MT Using Syntactic Guides

引用

International Joint Symposium on Artificial Intelligence and Natural language Processing (iSAI-NLP)

作者： Jinnawat Makwisai Prachya Boonkwan Sasiporn Usanavasin Natsuda Kaothanthong Manabu Okumura School of Information Computer and Communication Technology Sirindhorn International Institute of Technology Thammasat University Pathum Thani 12120 Thailand National Electronics and Computer Technology Center National Science and Technology Development Agency Pathum Thani 12120 Thailand School of Management Technology Sirindhorn International Institute of Technology Thammasat University Pathum Thani 12120 Thailand Precision and Intelligence Laboratory Tokyo Institute of Technology Yokohama 226-8503 Japan

This paper explores the profound impact of incorporating dependency tree information into the fine-tuning process of Transformer-based machine translation (MT) systems. The study investigates two translation scenarios: from multilingual source languages, including Thai, Japanese, and Chinese, to English as the target, and from English as the source language to these same multilingual target languages. The empirical results of this analysis clearly demonstrate that leveraging dependency trees in the target language significantly improves translation quality, leading to a substantial 50% improvement in validation scores compared to models trained without dependency tree information. Moreover, the inclusion of dependency trees notably enhances translation performance across different language pairs, except for cases involving word pre-segmented source and target datasets without dependency tree annotations. These findings suggest that utilizing dependency trees during the fine-tuning process can yield superior translation quality while enabling the use of fewer training data entries, thereby optimizing the MT system’s efficiency and effectiveness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Framework of Computer-Based Learning System Based on Self-Regulated Model in English Writing 14

A Framework of Computer-Based Learning System Based on Self-...

引用

14th International Joint Symposium on Artificial Intelligence and Natural language Processing, iSAI-NLP 2019

作者： Phodong, Kanyalag Supnithi, Thepchai Kongkachandra, Rachada Thammasat University Department of Computer Science Pathumthani Thailand Language and Semantic Technology Laboratory Intelligent Informatics Research Unit National Electronic and Computer Technology Center Pathumthani Thailand

ISBN: (纸本)9781728156316

This paper presents a design phase of a computer based learning system for English writing in Thai EFL learners. This system is designed to incorporate the self-regulated model and set the components of linguistics and machine translation as a learning environment. The system is designed based on three main phases of self-regulated model: forethought phase, performance phase, and self-reflection phase. The learning environment used to guide completely target sentence writing. Moreover, the display of user interface is designed for using as assisting tool for supporting a student's self-regulated learning in English writing. There are three main modules of the system that consist of learning profile acquisition, learning behavior collection, and learning analytics. The system design is an important phase to encourages action between learners and computer-based learning system for learning in English writing. Then, learners' behavior are collected into data logs store for learning analysis. This system aims to collect Thai EFL learners' behavior and find the behavioral pattern that could helpful reference for improve system and teaching materials in the future. © 2019 IEEE.

关键词： User interfaces

来源：评论

学校读者我要写书评

暂无评论

A Study on an Effect of Using Deep Learning in Thai-English Machine Translation Processes 14

A Study on an Effect of Using Deep Learning in Thai-English ...

引用

14th International Joint Symposium on Artificial Intelligence and Natural language Processing, iSAI-NLP 2019

作者： Luekhong, Prasert Limkonchotiwat, Peerat Ruangrajitpakorn, Taneth College of Integrated Science and Technology Rajamangala University of Technology Lanna Chiang Mai Thailand Rajamangala University of Technology Department of Computer Engineering Faculty of Engineering Lanna Chiang Mai Thailand Language and Semantic Technology Laboratory National Electronics and Computer Technology Center Pathumthani Thailand

ISBN: (纸本)9781728156316

Deep learning has been used in many fields including natural language processing. This paper aims to study the effect of applying deep learning in machine translation processes including word segmentation and translation model generation. We compare the results of the process from traditional statistical method and deep learning and analyze the difference. From experiment, the results indicated that the processes from deep learning obtained higher score in overall. Word segmentation from Bidirectional neural network yielded 0.861 f1 score which was higher than standard n-gram based system for 0.081. The translation results within dataset show that the neural-network-based translation got the best BLEU score in average for 0.43 in which are higher than the traditional statistical approach for 0.16. The result analysis indicates that the neural-network-based translation can translate better for Thai sentences containing unknown words and those with numerical classifier expression. © 2019 IEEE.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

A Learning Model to Improve Learning Outcome on Experiential Learning in a Multi-Phase Internship: A Case Study of the Internship Program of a Thai University. 8

A Learning Model to Improve Learning Outcome on Experiential...

引用

8th IIAI International Congress on Advanced Applied Informatics, IIAI-AAI 2019

作者： Kingkaew, Chanakarn Supnithi, Thepchai Theeramunkong, Thanaruk Morita, Kai Tanaka, Koji Ikeda, Mitsuru School of Information Computer and Communication Technology Sirindhorn International Institute of Technology Thammasat University Pathum Thani Thailand Language and Semantic Technology Laboratory National Electronics and Computer Technology Center Pathum Thani Thailand School of Knowledge Science Japan Advanced Institute of Science and Technology Nomi Japan Department of Psychological Science College of Informatics and Human Communication Kanazawa Institute of Technology Nonoichi Japan

ISBN: (纸本)9781728126272

The internship is one of the crucial learning processes that provide a unique experience to combine what they learned in university courses and how to use it in the real society during their university lives as it provides the opportunity to learn through work experience offered by the external organization. Although learner can learn career skills through the internship, most learner cannot make good reflection and cannot learn how to learn from experience. We think the cause of this problem is that most learner cannot reflect their experience and cannot clear the abstract concept and apply it to the next trial. This implies that those learners can improve their learning through internship by acquiring experiential learning skills. In the internship in general, the mentor has the important education roles. One is to foster career skills learning, and the other is to promote experiential learning. However, in most case, the mentor is very good at career skill teaching, but most of the mentor is not well prepared to support experiential learning. In response to the above problems, to support learners to realized good learning, we provide a learning environment for the learner to support experiential learning. A guideline for the learner to let them aware of the critical concept of internship and experiential learning. Also, we provide a web system to practice reflective writing to understand and able to apply experiential learning concepts. In this paper, as the preliminary research to develop learning environment, the analysis of the learner learning activity of the current internship program and show the prototype of the conceptualization of the educational concepts to support the experiential learning. © 2019 IEEE.

关键词： Knowledge management

来源：评论

学校读者我要写书评

暂无评论

A Framework of Computer-Based Learning System Based on Self-Regulated Model in English Writing

A Framework of Computer-Based Learning System Based on Self-...

引用

International Joint Symposium on Artificial Intelligence and Natural language Processing (iSAI-NLP)

作者： Kanyalag Phodong Thepchai Supnithi Rachada Kongkachandra Department of Computer Science Thammasat University Pathumthani Thailand Language and Semantic Technology Laboratory Intelligent Informatics Research Unit National Electronic and Computer Technology Center Pathumthani Thailand

ISBN: (数字)9781728156316

ISBN: (纸本)9781728156323

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Study on an Effect of Using Deep Learning in Thai-English Machine Translation Processes

A Study on an Effect of Using Deep Learning in Thai-English ...

引用

International Joint Symposium on Artificial Intelligence and Natural language Processing (iSAI-NLP)

作者： Prasert Luekhong Peerat Limkonchotiwat Taneth Ruangrajitpakorn College of Integrated Science and Technology Rajamangala University of Technology Lanna Chiang Mai Thailand Department of Computer Engineering Faculty of Engineering Rajamangala University of Technology Lanna Chiang Mai Thailand Language and Semantic Technology Laboratory National Electronics and Computer Technology Center Pathumthani Thailand

ISBN: (数字)9781728156316

ISBN: (纸本)9781728156323

关键词：

来源：评论

学校读者我要写书评

暂无评论

Correction: The cognitive‑linguistic profiles and academic performances of Chinese children with dyslexia across cultures: Beijing, Hong Kong, and Taipei

引用

Annals of dyslexia 2024年第2期74卷 243页

作者： Dora Jue Pan Xiangzhi Meng Jun Ren Lee Melody Chi Yi Ng Catherine McBride School of Humanities and Social Science The Chinese University of Hong Kong Shenzhen China. School of Psychological and Cognitive Sciences Beijing Key Laboratory of Behavior and Mental Health Peking University Beijing China. mengxzh@***. Department of Educational Psychology and Counseling National Taiwan Normal University Taipei Taiwan. Chinese Language and Technology Center National Taiwan Normal University Taipei Taiwan. Department of Psychology The Chinese University of Hong Kong Hong Kong China. Department of Human Development and Family Science Purdue University West Lafayette USA. mcbridca@purdue.edu.

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：