• Journal of the Semiconductor & Display Technology
• /
• v.22 no.3
• /
• pp.142-148
• /
• 2023

In this paper, we identify performance issues in executing compute kernels from PolyBench, which includes compute kernels that are the core computational units of various data-intensive workloads, such as deep learning and data-intensive applications, on Processing-in-Memory (PIM) devices. Therefore, using our in-house simulator, we measured and compared the various performance metrics of workloads based on traditional out-of-order and in-order processors with Processing-in-Memory-based systems. As a result, the PIM-based system improves performance compared to other computing models due to the short-term data reuse characteristic of computational kernels from PolyBench. However, some kernels perform poorly in PIM-based systems without a multi-layer cache hierarchy due to some kernel's long-term data reuse characteristics. Hence, our evaluation and analysis results suggest that further research should consider dynamic and workload pattern adaptive approaches to overcome performance degradation from computational kernels with long-term data reuse characteristics and hidden data locality.

• Journal of KIISE:Software and Applications
• /
• v.36 no.8
• /
• pp.642-652
• /
• 2009

In order to improve the performance of the existing relation extraction approaches, we propose a method for combining two pivotal concepts which play an important role in classifying semantic relationships between entities in text. Having built a composite kernel-based relation extraction system, which incorporates both entity features and syntactic structured information of relation instances, we define nine classes of lexical features and synthetically apply them to the system. Evaluation on the ACE RDC corpus shows that our approach boosts the effectiveness of the existing composite kernels in relation extraction. It also confirms that by integrating the three important features (entity features, syntactic structures and contextual lexical features), we can improve the performance of a relation extraction process.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.6
• /
• pp.38-44
• /
• 2019

A Convolution Neural Network(CNN) model was utilized to detect surface cracks in asphalt concrete pavements. The CNN used for this study consists of five layers with 3×3 convolution filter and 2×2 pooling kernel. Pavement surface crack images collected by automated road surveying equipment was used for the training and testing of the CNN. The performance of the CNN was evaluated using the accuracy, precision, recall, missing rate, and over rate of the surface crack detection. The CNN trained with the largest amount of data shows more than 96.6% of the accuracy, precision, and recall as well as less than 3.4% of the missing rate and the over rate.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.7
• /
• pp.374-379
• /
• 2016

On-line process monitoring schemes are designed to give early warnings of process faults. In the artificial intelligence and machine learning fields, reliable approaches have been utilized, such as kernel-based nonlinear techniques. This work presents a kernel-based empirical monitoring scheme with a small sample problem. The measurement data of normal operations are easy to collect, whereas special events or faults data are difficult to collect. In such situations, noise filtering techniques can be helpful in enhancing the process monitoring performance. This can be achieved by the preprocessing of raw process data and eliminating unwanted variations of data. In this work, the performance of several monitoring schemes was demonstrated using three-dimensional batch process data. The results showed that the monitoring performance was improved significantly in terms of the detection success rate.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.339-346
• /
• 2003

This paper proposes a multiple classifier system having massive micro classifiers. The micro classifiers are trained by using a local set of training patterns. The k nearest neighboring training patterns of one training pattern comprise the local region for training a micro classifier. Each training pattern is incorporated with one or more micro classifiers. Two types of micro classifiers are adapted in this paper. SVM with linear kernel and SVM with RBF kernel. Classification is done by selecting the best micro classifier among the micro classifiers in vicinity of incoming test pattern. To measure the goodness of each micro classifier, the weighted sum of correctly classified training patterns in vicinity of the test pattern is used. Experiments have been done on Elena database. Results show that the proposed method gives better classification accuracy than any conventional classifiers like SVM, k-NN and the conventional classifier combination/selection scheme.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.133-138
• /
• 2020

This paper introduces a method for improving accuracy using the first convolution layer, which is not used in most modified CNN(: Convolution Neural Networks). In CNN, such as GoogLeNet and DenseNet, the first convolution layer uses only the traditional methods(3×3 convolutional computation, batch normalization, and activation functions), replacing this with RGB-csb. In addition to the results of preceding studies that can improve accuracy by applying RGB values to feature maps, the accuracy is compared with existing CNN using a limited number of images. The method proposed in this paper shows that the smaller the number of images, the greater the learning accuracy deviation, the more unstable, but the higher the accuracy on average compared to the existing CNN. As the number of images increases, the difference in accuracy between the existing CNN and the proposed method decreases, and the proposed method does not seem to have a significant effect.

• Journal of the Korea Convergence Society
• /
• v.11 no.3
• /
• pp.19-26
• /
• 2020

In this paper, we propose a method to analyze the loss region of the dictionary-based super resolution result learned for image quality improvement and to map the learning data according to the analyzed loss region. In the conventional learned dictionary-based method, a result different from the feature configuration of the input image may be generated according to the learning image, and an unintended artifact may occur. The proposed method estimate loss information of low resolution images by analyzing the reconstructed contents to reduce inconsistent feature composition and unintended artifacts in the example-based super resolution process. By mapping the training data according to the final interpolation feature map, which improves the noise and pixel imbalance of the estimated loss information using a Gaussian-based kernel, it generates super resolution with improved noise, artifacts, and staircase compared to the existing super resolution. For the evaluation, the results of the existing super resolution generation algorithms and the proposed method are compared with the high-definition image, which is 4% better in the PSNR (Peak Signal to Noise Ratio) and 3% in the SSIM (Structural SIMilarity Index).

• Environmental and Resource Economics Review
• /
• v.28 no.4
• /
• pp.467-495
• /
• 2019

Uncertainty of renewable energy such as photovoltaic(PV) power is detrimental to the flexibility of the power system. Therefore, precise prediction of PV power generation is important to make the power system stable. The purpose of this study is to forecast PV power generation using meteorological data including particulate matter(PM). In this study, PV power generation is predicted by support vector machine using RBF kernel function based on machine learning. Comparing the forecasting performances by including or excluding PM variable in predictor variables, we find that the forecasting model considering PM is better. Forecasting models considering PM variable show error reduction of 1.43%, 3.60%, and 3.88% in forecasting power generation between 6am~8pm, between 12pm~2pm, and at 1pm, respectively. Especially, the accuracy of the forecasting model including PM variable is increased in daytime when PV power generation is high.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.581-584
• /
• 2019

Food consumption is growing worldwide every year owing to a growing population. Hence, the increasing population needs the production of sufficient and good quality food products. Strawberry is one of the world's most famous fruit. To obtain the highest strawberry output, we worked with three strawberry varieties supplied with three kinds of nutrient water in a greenhouse and with the outcome of the strawberry production, the highest yielding strawberry variety is detected. This Study uses the nutrient water consumed every day by the highest yielding strawberry variety. The atmospheric temperature, humidity and CO2 levels within the greenhouse are identified and used for the prediction, since the water consumption by any plant depends primarily on weather conditions. Machine learning techniques show successful outcomes in a multitude of issues including time series and regression issues. In this study, daily nutrient water consumption of strawberry plants is predicted using machine learning algorithms is proposed. Four Machine learning algorithms are used such as Linear Regression (LR), K nearest neighbour (KNN), Support Vector Machine with Radial Kernel (SVM) and Gradient Boosting Machine (GBM). Gradient Boosting System produces the best results.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1567-1577
• /
• 2020

Urban land cover classification is role in urban planning and management. So, it's important to improve classification accuracy on urban location. In this paper, machine learning model, Support Vector Machine (SVM) and Artificial Neural Network (ANN) are proposed for urban land cover classification based on high resolution satellite imagery (KOMPSAT-3A). Satellite image was trained based on 25 m rectangle grid to create training data, and training models used for classifying test area. During the validation process, we presented confusion matrix for each result with 250 Ground Truth Points (GTP). Of the four SVM kernels and the two activation functions ANN, the SVM Polynomial kernel model had the highest accuracy of 86%. In the process of comparing the SVM and ANN using GTP, the SVM model was more effective than the ANN model for KOMPSAT-3A classification. Among the four classes (building, road, vegetation, and bare-soil), building class showed the lowest classification accuracy due to the shadow caused by the high rise building.