Article Per Year (5 Year)
p-Index From 2020 - 2025
0.61
P-Index
This Author published in this journals
All Journal JOIV : International Journal on Informatics Visualization
Ng, Kok Why
Unknown Affiliation
Computer Science & IT

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

 1 
Drowsiness Detection System Through Eye and Mouth Analysis Belle Lim, Bey-Ee; Ng, Kok Why; Ng, Sew Lai
JOIV : International Journal on Informatics Visualization Vol 7, No 4 (2023)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.7.4.2288

Abstract

Traffic jams are one of the serious issues in many developed countries. After the pandemic, many employees were allowed to travel interstate to work. This contributes to more severe jams, especially in the capital and nearby states. Long-distance driving and congestion can easily make the drivers sleepy and thus lead to traffic accidents. This paper aims to study and analyze facial cues to detect early symptoms of drowsy driving. The proposed method employs a deep learning approach, utilizing ensemble CNNs and Dlib's 68 landmark face detectors to analyze the facial cues. The analyzed symptoms include the frequency of eyes opened or closed and yawning or no yawning. Three individual CNN models and an ensemble CNN structure are built for the classification of the eyes and mouth yawn. The model training and validation accuracy graph and training loss and validation loss graph are plotted to verify that the models have not been overfitted. The ensemble CNN models achieved an approximate accuracy of 97.4% from the eyes and 96.5% from the mouth. It outperforms the other pre-trained models. The proposed system can immediately alert the driver and send text drowsy messages and emails to the third party, ensuring timely intervention to prevent accidents. The proposed method can be integrated into vehicles and transportation systems to ensure driver's safety. It can also be applied to monitor the driving behavior of those who drive long distances
Processing Plant Diseases Using Transformer Model Marcus Lye, Hong Zheng; Ng, Kok Why
JOIV : International Journal on Informatics Visualization Vol 7, No 4 (2023)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.7.4.2291

Abstract

Agriculture faces challenges in achieving high-yield production while minimizing the use of chemicals. The excessive use of chemicals in agriculture poses many problems. Accurate disease diagnosis is crucial for effective plant disease detection and treatment. Automatic identification of plant diseases using computer vision techniques offers new and efficient approaches compared to traditional methods. Transformers, a type of deep learning model, have shown great promise in computer vision, but as the technology is still new, many vision transformer models struggle to identify diseases by examining the entire leaf. This paper aims to utilize the vision transformer model in analyzing and identifying common diseases that hinder the growth and development of plants through the plant leave images. Besides, it aims to improve the model's stability by focusing more on the entire leaf than individual parts and generalizing better results on leaves not in the image center. Added features such as Shift Patch Tokenization, Locality Self Attention, and Positional Encoding help focus on the whole leaf. The final test accuracy obtained is 89.58%, with relatively slight variances in precision, accuracy, and F1 score across classes, as well as satisfactory model robustness towards changes in leaf orientation and position within the image. The model's effectiveness shows the vision transformer's potential for automated plant disease diagnosis, which can help farmers take timely measures to prevent losses and ensure food security.
Continuous Training of Recommendation System for Airbnb Listings Using Graph Learning Chan, Yun Hong; Ng, Kok Why; Haw, Su Cheng; Palanichamy, Naveen
JOIV : International Journal on Informatics Visualization Vol 8, No 1 (2024)
Publisher : Society of Visual Informatics

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62527/joiv.8.1.2315

Abstract

Recommender systems are getting increasingly important nowadays as they can boost user engagement and benefit businesses. However, there remain some unsolved problems. This paper will address two key performance issues. First, the limited ability to identify and leverage intrinsic relationships between data points. Second, the inability to adapt to new data. The first issue is proposed to be addressed through a Graph Neural Network (GNN) to curate better recommendations. GNN will be trained with Airbnb’s review data to utilize its outstanding expressive power to represent complex user-listing interactions at scale, followed by generating embeddings to compute the relevant recommendations to the users. With the generated embeddings, the recommender system will compute a recommendation list to every user based on the embedding similarity between the user and listings or the user’s first-ever reviewed listing and listings. The second issue is proposed to be resolved by incorporating Continuous Training. The proposed recommender system employs GraphSAGE with a customized Rating-Weighted Triplet Ranking Loss function, which outperformed unsupervised GraphSAGE. Offline simulation validated the recommender system's ability to learn from the latest data and improve over time. Overall, the proposed user-to-item (U2I) recommendation rating-weighted GraphSAGE substantially increased by 99.88% in hit-rate@5 and 98.15% in coverage. This offers an effective solution for enhancing the recommender system for Airbnb listings. This research validates the efficacy of GNN-based recommendations in capturing user-item relationships to aid in predicting relevant recommendations, thus significantly driving up the adoption of GNN-based recommender systems.
Co-Authors Belle Lim, Bey-Ee Chan, Yun Hong Haw, Su Cheng Marcus Lye, Hong Zheng Ng, Sew Lai Palanichamy, Naveen