The Joint Symposium on Computational Intelligence (JSCI) is an event that was first organized in 2016. The event was initiated by the IEEE Computational Intelligence Society Thailand Chapter (IEEE-CIS Thailand Chapter), which aims to support research students and young researchers to create a place that enables participants to share and discuss their research before publishing their works. The event is open to all researchers who want to broaden their knowledge of computational intelligence. The symposium will feature student paper presentations as well as invited talks.
JSCI is a rehearsal platform for students who want to extend their work to present and publish at Scopus conference proceedings or submit for Scopus-Indexed Journals. JSCI is a stage for undergraduate and graduate students to practice their research work under the supervision of professors in the community. It is an activity organized by the IEEE Computational Intelligence Society Thailand Chapter. Students who participated in this event will gain some experience in both the presentation and written academic papers for conferences and journals.
This is a very good opportunity to share and discuss your work with other researchers in computational intelligence. If you are interested in presenting your work at this symposium, please submit a paper at https://easychair.org/conferences/?conf=jsci15. The paper must conform to the standard of IEEE Manuscript Templates for Conference Proceedings, which is available to download at https://www.ieee.org/conferences/publishing/templates.html or https://www.overleaf.com/latex/templates/ieee-conference-template/grfzhhncsfqn. A 4-6-page paper must be submitted to be reviewed. The papers will be double-blind peer-reviewed (please do not include authors’ names for the first-round submission).
Areas of Interest:
Key Areas of Focus:
From the depth of deep learning advancements and the natural selection-inspired algorithms of evolutionary computation to the nuanced decision-making capabilities of fuzzy systems and the integrated strength of hybrid intelligent systems, this collective exploration is not only solving complex problems across various sectors but also charting the future course of AI development. The challenges and opportunities within these areas underscore the ongoing quest for algorithmic innovation, scalability, and ethical considerations in AI technologies.
1. Neural Networks:
• Topics: Advances in Deep Learning, CNNs, RNNs, Generative AI, Attention Mechanisms, Transformers, Large Language Models, Multi-modal Models, and Explainable AI (XAI).
• Challenges: Pioneering new applications, expanding the horizons of AI technology, and addressing ethical and societal implications.
2. Evolutionary Computation:
• Topics: Biological Evolution, Genetic Algorithms, Genetic Programming, Evolutionary Strategies, Metaheuristics, and Swarm Intelligence.
• Challenges: Algorithmic innovation, uncovering new application domains, and enhancing the scalability and efficiency of evolutionary approaches.
3. Fuzzy Systems:
• Topics: Cutting-edge fuzzy clustering, fuzzy neural networks, type-2 fuzzy systems, and applications in big data, IoT, and decision-making under uncertainty.
• Challenges: Managing large datasets, synergizing with AI technologies, enhancing fuzzy system transparency, and developing robust fuzzy systems for real-world applications.
4. Hybrid Intelligent Systems:
• Topics: Fusion of neural networks, evolutionary algorithms, fuzzy logic, and other CI techniques; innovative use cases in complex systems such as autonomous vehicles, smart cities, and personalized healthcare.
• Challenges: Crafting robust, adaptable systems, balancing algorithmic trade-offs, and designing efficient architectures for real-time performance.
Submission Details:
• Abstracts should highlight the novelty, impact, and practical relevance of the work in Computational Intelligence.
• Length: 4-6 pages, inclusive of figures, tables, and references.
• Format: Must comply with IEEE conference formatting standards.
• Deadline: All submissions must be received by April 7, 2024.
Review and Publication:
• Each paper will undergo a thorough review by a panel of international experts, focusing on originality, technical quality, clarity of presentation, and thematic relevance.
• Excepted/Presented papers will be extended for Scopus indexed conference proceedings or extension and publication in Scopus indexed journals.
Important Dates
• Submission Deadline: April 7, 2024
• Acceptance Notification: April 21, 2024
• Final Paper Submission: May 5, 2024
• Symposium Dates: May 17-18, 2024
Your contribution is eagerly anticipated, as it will enrich the ongoing dialogue and shape the future direction of this dynamic field. We look forward to welcoming you to KMUTNB for JSCI15 and to fostering a stimulating and rewarding experience for all participants.
Advisory
• Chanboon Sathitwiriyawong (King Mongkut’s Institute of Technology Ladkrabang)
• Jonathan H. Chan (King Mongkut’s University of Technology Thonburi)
Chair
• Phayung Meesad (King Mongkut’s University of Technology North Bangkok)
Organizing Committee
• Bunthit Watanapa (King Mongkut’s University of Technology Thonburi)
• Kitsuchart Pasupa (King Mongkut’s Institute of Technology Ladkrabang)
• Kuntpong Woraratpanya (King Mongkut’s Institute of Technology Ladkrabang)
• Maleerat Maliyaem (King Mongkut’s University of Technology North Bangkok)
• Nat Dilokthanakul (King Mongkut’s Institute of Technology Ladkrabang)
• Pornchai Mongkolnam (King Mongkut’s University of Technology Thonburi)
• Pramuk Boonsieng (Thai-Nichi Institute of Technology)
• Saichon Jaiyen (King Mongkut’s University of Technology Thonburi)
• Sansanee Auephanwiriyakul (Chiang Mai University)
• Sarayut Nonsiri (Thai-Nichi Institute of Technology)
• Soradech Krootjohn (King Mongkut’s University of Technology North Bangkok)
SCHEDULE ON May 17, 2024
Date/Time: May 17, 2024, 1:00-5:30pm (Thailand Time)
Value: Navamintararachinee Building, KMUTNB
Google Meet: meet.google.com/dui-ninh-cdt
Chair: Phayung Meesad
| Time | Speaker | Title | Abstract |
| 1.00pm-3.00pm | Prof. Dr.-Ing. habil. Dr. h.c. Herwig Unger, Fern University in Hagen, Germany | Brain-inspired Methods for Big Data Computing and Learning | After a brief introduction, we delve into the foundational structures and functional principles of the brain, illuminating the human brain’s role as a predictive engine that operates on sequence learning and recognition mechanisms. Expanding upon these revelations, we unveil the essential framework of the ‘GraphLearner,’ a model meticulously designed to embody these core principles. Initial experiment results showcase its effectiveness, particularly within the realm of Natural Language Processing (NLP). Furthermore, the presentation explores how the GraphLearner can bolster attention, parallel processing, and hierarchy formation. The final segment of the talk focuses on the essential approximations necessary for recognizing similar sequences and ensuring the fault tolerance of the GraphLearner. This involves the integration of ‘Sparse Distributed Representations,’ a concept thoroughly dissected after a concise introduction. |
| 3:00pm-3:15pm | Coffee Break | ||
| 3:15pm-3:30pm | Nguyen Minh Tuan, Faculty of Applied Science, KMUTNB, Thailand | On Students’ Behavior Prediction for Library Service Quality Using Bidirectional Deep Machine Learning, | Library service quality has been taken into account after the COVID-19 pandemic to propose appropriate conditions under the rapid change of technology circumstances. The main study in this paper is to consider the evaluation of students’ sentiments to understand the role of library service quality after the pandemic outbreak and evaluate library service quality consistent with the new situation. For this study, we employ deep learning models such as Convolutional Bidirectional Long Short-Term Memory (Conv-BiLSTM) and Convolutional Bidirectional Gated Recurrent Unit (Conv-BiGRU), Attention and Transformer TF BERT model. Our findings indicate that the Conv-BiLSTM (94.59%) and Conv-BiGRU (94.33%) outperformed the others, achieving the high est accuracy for predicting Vietnamese students’ sentiments about library service quality. |
| 3:30pm-3:45pm | Warameth Nuipian, Faculty of Information Technology and Digital Innovation, KMUTNB, Thailand | A Comprehensive Case Study on Deep Reinforcement Learning for Stock Portfolio Optimization, Warameth Nuipian, Phayung Meesad, and Maleerat Maliyaem. | This paper presents a comprehensive case study on the application of Deep Q-Networks (DQN), a deep reinforcement learning (DRL) algorithm, to the problem of stock portfolio optimization. Traditional methods often struggle to adapt to the complex and dynamic nature of financial markets. In contrast, DRL enables the automatic extraction of features from data and the ability to learn optimal strategies through interaction with the environment. The key components of the DRL framework for optimization are described, including the environment, agent, state, action, observation, reward, and Q-values. The paper demonstrates the implementation of DQN using Python and compares its performance to traditional techniques. Through this analysis, the potential of DRL to revolutionize quantitative finance. The strengths and limitations of the approach are discussed, along with future research directions. The results highlight the effectiveness of DQN in generating competitive returns, managing risk, adapting to market conditions, and providing valuable decision support for portfolio managers. This case study underscores the transformative potential of integrating DRL into optimization for data-driven, adaptive, and automated investment. |
| 3:45pm-4:00pm | Thasorn Chalongvorachai, Faculty of Information Technogloy, KMITL, Thailand | An Application of Latent Space Representation for Signal Imputation | Addressing missing data within anomalous events is a pervasive challenge in data analysis, hindering both the completeness of information and the efficacy of imputation methods. Traditional approaches, spanning from statistical methods to machine learning techniques, often struggle to adequately restore original data characteristics in the presence of substantial data loss. To tackle these issues, we propose a novel framework called “Latent Space Representation for Signal Imputation.” This framework harnesses the power of a variational autoencoder (VAE) to learn from complete data and derive a latent space representation. By extracting coordination points from this latent space, we establish a robust prediction model for data imputation. Through experimentation on anomalous signals from a water treatment testbed, our method surpasses baseline techniques, notably enhancing the similarity of output signals even under conditions of severe data loss. This success stems from the VAE decoder’s proficiency in data restoration and the framework’s ability to discern relationships among individual signals. Our approach offers a promising avenue for addressing missing data challenges within anomalous event contexts. |
| 4:00pm-4:15pm | Matheus Emerick de Magalhães, Instituto Federal do Piauí, Brazil | A proposal for an Application Programming Interface (API) based on Deep Learning for Disease Detection in Tomato Leaves, Matheus Emerick de Magalhães, Carlos Estevão Bastos Sousa, Jose Guilherme Mouta Ferreira, Jurandir Cavalcante Lacerda Júnior, Carlos Eduardo Barbosa and Jano Moreira de Souza | Tomatoes are a rich source of essential nutrients, such as vitamin C and lycopene, playing a crucial role in the population’s diet and the local economy. This study arises from the need to address the specific challenges in growing these fruits, focusing on identifying foliar diseases through deep learning models. Using a database with 11,000 images of tomato leaves, several classes of pathologies were explored, including bacterial spot, early blight, late blight, leaf mold, septorial leaf, spider mites, target spot, Mosaic virus, yellow leaves, and in addition to healthy samples. Three models were trained: ResNet50, MobileNet V2, and Inception V3. The results reveal that Inception V3 stands out, achieving an accuracy of 98 %, followed by ResNet50 with 96 % and MobileNet V2 with 95 %. Based on the best model, an API was developed, capable of processing images and offering information about control practices and methods. The tests validated the effectiveness of the API, representing a significant contribution to advances in disease detection in tomato plantations. |
| 4:15pm-4:30pm | Carlos Eduardo Guedes, Instituto Federal do Piauí, Brazil | Strengthening School Security: Threat Mitigation Through Computer Vision, Carlos Eduardo Guedes, Carlos Estevão Bastos Sousa, Matheus Emerick de Magalhães, Carlos Eduardo Barbosa and Jano Moreira de Souza, Brazil | |
| 4:30pm-4:45pm | |||
| 4:45pm-5:30pm | IEEE CIS Chapter Meeting | ||
| 5:30pm-8:30pm | Dinner |
Contact and Additional Information:
• Email: phayung.m@itd.kmutnb.ac.th
• Symposium Website: https://site.ieee.org/thailand-cis/event/jsci15
• Venue: King Mongkut’s University of Technology North Bangkok, Thailand
• Paper Submission: https://easychair.org/conferences/?conf=jsci15
Don’t miss this opportunity to be part of JSCI15 and contribute to shaping the future of Computational Intelligence. We eagerly await your submissions and look forward to welcoming you to Bangkok in May 2024!