• Management & Information Systems Review
• /
• v.36 no.4
• /
• pp.33-51
• /
• 2017

With the advent of 'The Forth Industrial Revolution' and the growing demand for quality of life due to economic growth, needs for the quality of medical services are increasing. Artificial intelligence has been introduced in the medical field, but it is rarely used in chronic skin diseases that directly affect the quality of life. Also, atopic dermatitis, a representative disease among chronic skin diseases, has a disadvantage in that it is difficult to make an objective diagnosis of the severity of lesions. The aim of this study is to establish an intelligent severity recognition model of atopic dermatitis for improving the quality of patient's life. For this, the following steps were performed. First, image data of patients with atopic dermatitis were collected from the Catholic University of Korea Seoul Saint Mary's Hospital. Refinement and labeling were performed on the collected image data to obtain training and verification data that suitable for the objective intelligent atopic dermatitis severity recognition model. Second, learning and verification of various CNN algorithms are performed to select an image recognition algorithm that suitable for the objective intelligent atopic dermatitis severity recognition model. Experimental results showed that 'ResNet V1 101' and 'ResNet V2 50' were measured the highest performance with Erythema and Excoriation over 90% accuracy, and 'VGG-NET' was measured 89% accuracy lower than the two lesions due to lack of training data. The proposed methodology demonstrates that the image recognition algorithm has high performance not only in the field of object recognition but also in the medical field requiring expert knowledge. In addition, this study is expected to be highly applicable in the field of atopic dermatitis due to it uses image data of actual atopic dermatitis patients.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• Journal of the Korean earth science society
• /
• v.40 no.5
• /
• pp.447-463
• /
• 2019

Sea surface wind is a fundamental element for understanding the oceanic phenomena and for analyzing changes of the Earth environment caused by global warming. Global research institutes have developed and operated scatterometers to accurately and continuously observe the sea surface wind, with the accuracy of approximately ${\pm}20^{\circ}$ for wind direction and ${\pm}2m\;s^{-1}$ for wind speed. Given that the spatial resolution of the scatterometer is 12.5-25.0 km, the applicability of the data to the coastal area is limited due to complicated coastal lines and many islands around the Korean Peninsula. In contrast, Synthetic Aperture Radar (SAR), one of microwave sensors, is an all-weather instrument, which enables us to retrieve sea surface wind with high resolution (<1 km) and compensate the sparse resolution of the scatterometer. In this study, we investigated the Geophysical Model Functions (GMF), which are the algorithms for retrieval of sea surface wind speed from the SAR data depending on each band such as C-, L-, or X-band radar. We reviewed in the simulation of the backscattering coefficients for relative wind direction, incidence angle, and wind speed by applying LMOD, CMOD, and XMOD model functions, and analyzed the characteristics of each GMF. We investigated previous studies about the validation of wind speed from the SAR data using these GMFs. The accuracy of sea surface wind from SAR data changed with respect to observation mode, GMF type, reference data for validation, preprocessing method, and the method for calculation of relative wind direction. It is expected that this study contributes to the potential users of SAR images who retrieve wind speeds from SAR data at the coastal region around the Korean Peninsula.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.2
• /
• pp.80-98
• /
• 2023

This study explores the effectiveness of various data preprocessing algorithms for improving subseasonal to seasonal (S2S) climate predictions from six climate forecast models and their Multi-Model Ensemble (MME) using a deep learning-based postprocessing model. A pipeline of data transformation algorithms was constructed to convert raw S2S prediction data into the training data processed with several statistical distribution. A dimensionality reduction algorithm for selecting features through rankings of correlation coefficients between the observed and the input data. The training model in the study was designed with TimeDistributed wrapper applied to all convolutional layers of U-Net: The TimeDistributed wrapper allows a U-Net convolutional layer to be directly applied to 5-dimensional time series data while maintaining the time axis of data, but every input should be at least 3D in U-Net. We found that Robust and Standard transformation algorithms are most suitable for improving S2S predictions. The dimensionality reduction based on feature selections did not significantly improve predictions of daily precipitation for six climate models and even worsened predictions of daily maximum and minimum temperatures. While deep learning-based postprocessing was also improved MME S2S precipitation predictions, it did not have a significant effect on temperature predictions, particularly for the lead time of weeks 1 and 2. Further research is needed to develop an optimal deep learning model for improving S2S temperature predictions by testing various models and parameters.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_3
• /
• pp.979-995
• /
• 2023

As wildfires are difficult to predict, real-time monitoring is crucial for a timely response. Geostationary satellite images are very useful for active fire detection because they can monitor a vast area with high temporal resolution (e.g., 2 min). Existing satellite-based active fire detection algorithms detect thermal outliers using threshold values based on the statistical analysis of brightness temperature. However, the difficulty in establishing suitable thresholds for such threshold-based methods hinders their ability to detect fires with low intensity and achieve generalized performance. In light of these challenges, machine learning has emerged as a potential-solution. Until now, relatively simple techniques such as random forest, Vanilla convolutional neural network (CNN), and U-net have been applied for active fire detection. Therefore, this study proposed an active fire detection algorithm using state-of-the-art (SOTA) deep learning techniques using data from the Advanced Himawari Imager and evaluated it over East Asia and Australia. The SOTA model was developed by applying EfficientNet and lion optimizer, and the results were compared with the model using the Vanilla CNN structure. EfficientNet outperformed CNN with F1-scores of 0.88 and 0.83 in East Asia and Australia, respectively. The performance was better after using weighted loss, equal sampling, and image augmentation techniques to fix data imbalance issues compared to before the techniques were used, resulting in F1-scores of 0.92 in East Asia and 0.84 in Australia. It is anticipated that timely responses facilitated by the SOTA deep learning-based approach for active fire detection will effectively mitigate the damage caused by wildfires.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.6B
• /
• pp.531-539
• /
• 2011

It has been widely acknowledged that climate system associated with extreme rainfall events was difficult to understand and extreme rainfall simulation in climate model was more difficult. This study developed a new model for extracting rainfall filed associated with extreme events as a way to characterize large scale climate system. Main interests are to derive location, size and direction of the rainfall field and this study developed an algorithm to extract the above characteristics from global climate data set. This study mainly utilized specific humidity and wind vectors driven by NCEP reanalysis data to define the rainfall field. Geometric first and second moments have been extensively employed in defining the rainfall field in selected zone, and an ellipsoid based model were finally introduced. The proposed geometric moments based ellipsoid model works equally well with regularly and irregularly distributed synthetic grid data. Finally, the proposed model was applied to space-time real rainfall filed. It was found that location, size and direction of the rainfall field was successfully extracted.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.7
• /
• pp.419-426
• /
• 2012

The pitch searching is a vital process in a vocoder. Generally, the method of pitch searching is employed after highlighting the periodicity, where a correlation is identified with the signal by changing the interval of two pulses. When the correlation value reaches the peak, the pitch can be found by the pulse interval because it is the repetition interval with most striking period. However if the identified period happens to be one of half period, double period or triple period, this cannot be considered as the pitch period. Many methods were suggested to solve this problem. An inaccurate pitch could be obtained as well, when there is an interval where signal amplitude is not constant but varies abruptly in the frame. To solve this matter, searching the pitch by dividing a frame into various subframes is adopted, but too much calculation has to be followed while it leads the correct value. This paper suggests an algorithm to resolve these two problems. First, to search the pitch after advance correction of the signal energy level with an estimated overall energy change ratio in the frame before pitch search to reduce half period, double period and triple period is suggested. Second, to vary the number of subframes by predicting the amplitude change rate in the frame by the energy ratio obtained by the above-mentioned method is advised. If these two methods are applied, the pitch searching time can be reduced and the general pitch searching performance can be improved without affecting the sound quality in the synthesized signal.

• Journal of Life Science
• /
• v.30 no.9
• /
• pp.813-818
• /
• 2020

The archaeal clusters of orthologous genes (arCOG) algorithm, which identifies common genes among archaebacterial genomes, was used to identify conservative genes among 168 archaebacterial strains. The numbers of conserved orthologs were 14, 10, 9, and 8 arCOGs in 168, 167, 166, and 165 strains, respectively. Among 41 conserved arCOGs, 13 were related to function J (translation, ribosomal structure, and biogenesis), and 10 were related to function L (replication, recombination, and repair). Among the 14 conserved arCOGs in all 168 strains, 6 arCOGs of tRNA synthetase comprised the highest proportion. Of the remaining 8 arCOGs, 2 are involved in reactions with ribosomes, 2 for tRNA synthesis, 2 for DNA replication, and 2 for transcription. These results showed the importance of protein expression in archaea. For the classes or orders having 3 or more members, genomic analysis was performed by averaging the distance values of the conservative arCOGs. Classes Archaeoglobi and Thermoplasmata of the phylum Euryarchaeota showed the lowest and the highest average of distance value, respectively. This study can provides data necessary for basic scientific research and the development of antibacterial agents and tumor control.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.42 no.12
• /
• pp.45-52
• /
• 2005

A systolic array formed by interconnecting a set of identical data-processing cells in a uniform manner is a combination of an algorithm and a circuit that implements it, and is closely related conceptually to arithmetic pipeline. High-performance computation on a large array of cells has been an important feature of systolic array. To achieve even higher degree of concurrency, it is desirable to make cells of systolic array themselves systolic array as well. The structure of systolic array with its cells consisting of another systolic array is to be called super-systolic array. This paper proposes a scalable bit-level super-systolic amy which can be adopted in the VLSI design including regular interconnection and functional primitives that are typical for a systolic architecture. This architecture is focused on highly regular computational structures that avoids the need for a large number of global interconnection required in general VLSI implementation. A bit-level super-systolic FIR filter is selected as an example of bit-level super-systolic array. The derived bit-level super-systolic FIR filter has been modeled and simulated in RT level using VHDL, then synthesized using Synopsys Design Compiler based on Hynix $0.35{\mu}m$ cell library. Compared conventional word-level systolic array, the newly proposed bit-level super-systolic arrays are efficient when it comes to area and throughput.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.12 s.354
• /
• pp.23-32
• /
• 2006

Fast and accurate power bus design (FAPUD) method for multi-layers high-speed digital boards is devised for the power supply network design tool for accurate and precise high speed board. FAPUD is constructed, based on two main algorithms of the PBEC (Path Based Equivalent Circuit) model and the network synthesis method. The PBEC model exploits simple arithmetic expressions of the lumped 1-D circuit model from the electrical parameters of a 2-D power distribution network. The circuit level design based on PBEC is carried with the proposed regional approach. The circuit level design directly calculates and determines the size of on-chip decoupling capacitors, the size and the location of off-chip decoupling capacitors, and the effective inductances of the package power bus. As a design output, a lumped circuit model and a pre-layout of the power bus including a whole decoupling capacitors are obtained after processing FAPUD. In the tuning procedure, the board re-optimization considering simultaneous switching noise (SSN) added by I/O switching can be carried out because the I/O switching effect on a power supply noise can be estimated over the operation frequency range with the lumped circuit model. Furthermore, if a design changes or needs to be tuned, FAPUD can modify design by replacing decoupling capacitors without consuming other design resources. Finally, FAPUD is accurate compared with conventional PEEC-based design tools, and its design time is 10 times faster than that of conventional PEEC-based design tools.