In this study, we report the process variation effect (PVE) including the work function fluctuation (WKF) on the DC/AC characteristic fluctuation of stacked gate-all-around silicon complementary field-effect transisto...
详细信息
An accurate predictive model of temperature and humidity plays a vital role in many industrial processes that utilize a closed space such as in agriculture and building management. With the exceptional performance of ...
详细信息
An accurate predictive model of temperature and humidity plays a vital role in many industrial processes that utilize a closed space such as in agriculture and building management. With the exceptional performance of deep learning on time-series data, developing a predictive temperature and humidity model with deep learning is propitious. In this study, we demonstrated that deep learning models with multivariate time-series data produce remarkable performance for temperature and relative humidity prediction in a closed space. In detail, all deep learning models that we developed in this study achieve almost perfect performance with an R value over 0.99.
The Middle Eastern and North African region is highly reliant on the oil and gas industry. Subsequently, the need for pipeline inspection and fault diagnosis has become paramount. Current inspection methods rely on ma...
详细信息
A sugarcane yield of one plantation area depends on several independent variables. Practically it is challenging to predict accurately by using conventional methods. This study aims to develop a decision model based o...
详细信息
ISBN:
(数字)9798331519643
ISBN:
(纸本)9798331519650
A sugarcane yield of one plantation area depends on several independent variables. Practically it is challenging to predict accurately by using conventional methods. This study aims to develop a decision model based on a combination of fuzzy logic and object-oriented methods to predict sugarcane yield. The research is conducted in four main stages, employing object-oriented methods for model design and fuzzy logic for model construction. Object and activity diagrams are used for the object-oriented model design. The fuzzy membership functions employed are a combination of trapezoidal and triangular shapes. The resulting decision model can simulate 2,225 data from plantation areas in Indonesia. Based on the 10 examples of plantation area data in Indonesia, plantation number one obtained the largest sugarcane yield, which was 4.79%, with a similarity value of 0.90 (when compared to manual calculations as its ground truth). This similarity value is a higher value when compared to the average similarity value, which is 0.89.
The existence of fine-grained image classification supporting smart retail provides effectiveness in recognizing products with high similarity. However, the generic classification method performs poorly in identifying...
详细信息
ISBN:
(数字)9798350327472
ISBN:
(纸本)9798350327489
The existence of fine-grained image classification supporting smart retail provides effectiveness in recognizing products with high similarity. However, the generic classification method performs poorly in identifying products from a subordinate category. This paper aims to identify augmentation techniques to leverage the Vision Transformer (ViT) model to classify the fine-grained grocery product, which involves embedding patches and transformer encoders to extract the main features. First, we develop a fine-grained image dataset with ColorJitter, CutOut, and combining both augmentations. Secondly, we perform experiments and analysis of ViT size, embedded patch size and image size in the patch embedding process. Lastly, the ViT model are evaluated according to the image sizes 224, 384, and 512 in accuracy, loss, and confusion matrix. The highest accuracy was obtained at 0.9922. The ColorJitter and CutOut improved the confusion matrix in ViT-B/16 and ViT-L/16 with an image size of 384 and 512. The results show that both augmentations in the ViT model are able to distinguish fine-grained grocery products.
Email spam detection is crucial for ensuring a positive user experience and maintaining communication security. This study presents a novel spam detection approach leveraging Logistic Regression, optimized through hyp...
详细信息
ISBN:
(数字)9798350357509
ISBN:
(纸本)9798350357516
Email spam detection is crucial for ensuring a positive user experience and maintaining communication security. This study presents a novel spam detection approach leveraging Logistic Regression, optimized through hyperparameter tuning and enhanced with Explainable Artificial Intelligence. The proposed method is evaluated on a large-scale dataset of emails collected from a real-world spam detection system. Term Frequency-Inverse Document Frequency is utilized for feature extraction, converting email text into numerical representations. XAI introduces interpretability by highlighting critical features such as "pill," "PHP," and "Businessweek," which indicate spam classification patterns. Metrics such as precision, recall, and the confusion matrix are employed to assess the model’s performance, ensuring a balanced evaluation. Hyperparameter optimization using Grid Search CV achieves an optimal regularization parameter (C = 100) and maximum iterations (maxiter=100), resulting in an impressive F1 score of 98.76%, accuracy of 98.82%, precision of 98.62%, and recall of 98.90%.
The early stage detection of benign and malignant pulmonary nodules plays an important role in clinical diagnosis. The malignancy risk assessment is usually used to guide the doctor in identifying the cancer stage and...
详细信息
Skin cancer's increasing incidence rates necessitate advanced diagnostic tools. This research uses MobileNet architecture to develop an enhanced system for skin cancer detection. MobileNet's efficient CNN arch...
详细信息
ISBN:
(数字)9798350378511
ISBN:
(纸本)9798350378528
Skin cancer's increasing incidence rates necessitate advanced diagnostic tools. This research uses MobileNet architecture to develop an enhanced system for skin cancer detection. MobileNet's efficient CNN architecture, combined with Keras for model lifecycle management and TensorFlow for real-time inference, forms the foundation of this study. Using 10,015 dermatoscopic pictures from seven different classifications of skin cancer in the HAM10000 dataset, the methodology includes several key stages: image preprocessing to correct illumination and adjust resolution, data augmentation to balance the dataset, and model training using Transfer Learning. The MobileNet model was trained over 50 epochs with a comprehensive architecture incorporating multiple specialized layers. By combining large-scale data analysis and adaptive learning, this methodology demonstrates The revolutionary possibilities of AI and ML in improving skin cancer diagnostics and public health outcomes. Keywords Skin Cancer Detection, Dermatoscopic Images, Machine Learning, Deep Learning
computer vision has emerged as an important subject of study, with several practical applications in a wide range of domains. OpenCV, a widely used framework, has played an important role in allowing computer vision t...
详细信息
Despite the increasing computing power of shared memory systems with high core counts, parallel graph processing frameworks cannot exploit it effectively. The reason behind this is the inherent challenges in parallel ...
详细信息
ISBN:
(数字)9798331515966
ISBN:
(纸本)9798331515973
Despite the increasing computing power of shared memory systems with high core counts, parallel graph processing frameworks cannot exploit it effectively. The reason behind this is the inherent challenges in parallel graph algorithms, which are efficient management of dynamically created tasks and irregular data access patterns. In this paper, we categorize several popular design choices into three design dimensions: (i) execution mode, (ii) data access pattern, and (iii) work activation. We provide their high-level parallel implementations and analyze various implementations of three representative iterative graph algorithms by considering these design dimensions. To gain a better understanding of design choices, we examine their impacts on performance, communication, scalability, and work efficiency. We also investigate the communication characteristics of the design choices on two state-of-the-art shared-memory platforms by performing micro-architectural analysis. Our microarchitectural analysis reveals that a topology-driven, pull-based model gives up to $20 x$ better performance.
暂无评论