Neural networks (NNs) have proven a useful surrogate model for the design and optimization of high frequency structures including antennas. Black-box NNs are known to have scalability and accuracy problems as the dime...
详细信息
Accurate prediction of students’ graduation time is a significant challenge for academic institutions, especially in the context of optimizing educational outcomes and resource allocation. However, there is a researc...
详细信息
ISBN:
(数字)9798331517601
ISBN:
(纸本)9798331517618
Accurate prediction of students’ graduation time is a significant challenge for academic institutions, especially in the context of optimizing educational outcomes and resource allocation. However, there is a research gap in identifying which machine learning algorithms are best suited for this task, particularly in the electrical and informaticsengineeringdepartment. This study addresses this gap by evaluating the performance of various machine learning algorithms in predicting students’ graduation time. Several algorithms, including Logistic Regression (LR), Decision Tree (DT), K-Nearest Neighbors (KNN), Random Forest (RF), Support Vector Machines (SVM), and Naive Bayes, were applied to a dataset consisting of academic and demographic records of students from a university in Indonesia. The evaluation used performance metrics such as accuracy, precision, recall, and F1-score. The results demonstrate that LR, KNN, DT, RF, and SVM exhibited comparable accuracy rates of 74%, with a weighted average F1-score of 0.85, indicating these algorithms are effective in classifying data. In contrast, Naive Bayes, while showing superior speed with an execution time of 0.018322 seconds, achieved lower performance with an accuracy of only 39% and a weighted average F1-score of 0.44. These findings suggest that selecting an algorithm should balance the trade-off between accuracy and time efficiency. For scenarios where both are important, LR and DT are optimal choices, while Naive Bayes may be suitable for faster processing at the expense of accuracy.
Epileptic seizure detection has been a complex task due to the chaos and non-stationariness observed in the electroencephalogram (EEG) signals. Most existing EEG-based seizure detection algorithms are patient-dependen...
详细信息
Indoor Environmental Quality (IEQ) is undeniably a crucial factor that affects the occupants’ overall health and well-being. However, there is a significant lack of available actionable and measurable real-time metri...
详细信息
Resonant operation, exploiting high quality-factor planar inductors, has recently enabled gigahertz (GHz) applications for large-area electronics (LAE), providing a new technology platform for large-scale and flexible...
详细信息
A high reception of THz waves is important in the design of an antenna-coupled microbolometer, which is indicated by the ability to detect THz waves effectively from all directions. In this study, we implement the mea...
详细信息
Current solutions that rely on a single-server architecture have privacy, anonymity, integrity, and confidentiality limitations. Blockchain-based solutions can address some of these issues but face challenges regulati...
详细信息
Reliability, efficiency and continuity of power energy supplied is an area which receives increasing attention as the main infrastructure of power transmission and distribution systems in many countries is ageing. Hot...
详细信息
This paper presents the architecture of a Convolution Neural Network(CNN)accelerator based on a newprocessing element(PE)array called a diagonal cyclic array(DCA).As demonstrated,it can significantly reduce the burden...
详细信息
This paper presents the architecture of a Convolution Neural Network(CNN)accelerator based on a newprocessing element(PE)array called a diagonal cyclic array(DCA).As demonstrated,it can significantly reduce the burden of repeated memory accesses for feature data and weight parameters of the CNN models,which maximizes the data reuse rate and improve the computation ***,an integrated computation architecture has been implemented for the activation function,max-pooling,and activation function after convolution calculation,reducing the hardware *** evaluate the effectiveness of the proposed architecture,a CNN accelerator has been implemented for You Only Look Once version 2(YOLOv2)-Tiny consisting of 9 ***,the methodology to optimize the local buffer size with little sacrifice of inference speed is presented in this *** implemented the proposed CNN accelerator using a Xilinx Zynq ZCU102 Ultrascale+Field Programmable Gate Array(FPGA)and ISE Design *** FPGA implementation uses 34,336 Look Up Tables(LUTs),576 Digital Signal Processing(DSP)blocks,and an on-chip memory of only 58 KB,and it could achieve accuracies of 57.92% and 56.42% mean Average Precession@0.5 thresholds for intersection over union(mAP@0.5)using quantized 16-bit and 8-bit full integer data manipulation with only 0.68% as a loss for 8-bit version and computation time of 137.9 and 69 ms for each input image respectively using a clock speed of 200 *** speeds are expected to be doubled five times using a clock speed of 1GHz if implemented in a silicon System on Chip(SoC)using a sub-micron process.
This article introduces a novel Multi-agent path planning scheme based on Conflict Based Search (CBS) for heterogeneous holonomic and non-holonomic agents, designated as Heterogeneous CBS (HCBS). The proposed methodol...
详细信息
暂无评论