• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.29-39
• /
• 2021

As damages to individuals, private sectors, and businesses increase due to newly occurring cyber attacks, the underlying network security problem has emerged as a major problem in computer systems. Therefore, NIDS using machine learning and deep learning is being studied to improve the limitations that occur in the existing Network Intrusion Detection System. In this study, a deep learning-based NIDS model study is conducted using the Convolution Neural Network (CNN) algorithm. For the image classification-based CNN algorithm learning, a discrete algorithm for continuity variables was added in the preprocessing stage used previously, and the predicted variables were expressed in a linear relationship and converted into easy-to-interpret data. Finally, the network packet processed through the above process is mapped to a square matrix structure and converted into a pixel image. For the performance evaluation of the proposed model, NSL-KDD, a representative network packet data, was used, and accuracy, precision, recall, and f1-score were used as performance indicators. As a result of the experiment, the proposed model showed the highest performance with an accuracy of 85%, and the harmonic mean (F1-Score) of the R2L class with a small number of training samples was 71%, showing very good performance compared to other models.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Journal of Digital Convergence
• /
• v.19 no.7
• /
• pp.199-208
• /
• 2021

Automatic Post Editing(APE) is the study that automatically correcting errors included in the machine translated sentences. The goal of APE task is to generate error correcting models that improve translation quality, regardless of the translation system. For training these models, source sentence, machine translation, and post edit, which is manually edited by human translator, are utilized. Especially in the recent APE research, multilingual pretrained language models are being adopted, prior to the training by APE data. This study deals with multilingual pretrained language models adopted to the latest APE researches, and the specific application method for each APE study. Furthermore, based on the current research trend, we propose future research directions utilizing translation model or mBART model.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.42 no.2
• /
• pp.127-136
• /
• 2022

This study was conducted to estimate the damage of Whole Crop Maize (WCM) according to abnormal climate using machine learning and present the damage through mapping. The collected WCM data was 3,232. The climate data was collected from the Korea Meteorological Administration's meteorological data open portal. Deep Crossing is used for the machine learning model. The damage was calculated using climate data from the Automated Synoptic Observing System (95 sites) by machine learning. The damage was calculated by difference between the Dry matter yield (DMY)_normal and DMY_abnormal. The normal climate was set as the 40-year of climate data according to the year of WCM data (1978~2017). The level of abnormal climate was set as a multiple of the standard deviation applying the World Meteorological Organization(WMO) standard. The DMY_normal was ranged from 13,845~19,347 kg/ha. The damage of WCM was differed according to region and level of abnormal climate and ranged from -305 to 310, -54 to 89, and -610 to 813 kg/ha bnormal temperature, precipitation, and wind speed, respectively. The maximum damage was 310 kg/ha when the abnormal temperature was +2 level (+1.42 ℃), 89 kg/ha when the abnormal precipitation was -2 level (-0.12 mm) and 813 kg/ha when the abnormal wind speed was -2 level (-1.60 m/s). The damage calculated through the WMO method was presented as an mapping using QGIS. When calculating the damage of WCM due to abnormal climate, there was some blank area because there was no data. In order to calculate the damage of blank area, it would be possible to use the automatic weather system (AWS), which provides data from more sites than the automated synoptic observing system (ASOS).

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.43 no.1
• /
• pp.11-21
• /
• 2023

This study was conducted to estimate the damage of Whole Crop Corn (WCC; Zea Mays L.) according to abnormal climate using machine learning as the Representative Concentration Pathway (RCP) 4.5 and present the damage through mapping. The collected WCC data was 3,232. The climate data was collected from the Korea Meteorological Administration's meteorological data open portal. The machine learning model used DeepCrossing. The damage was calculated using climate data from the automated synoptic observing system (ASOS, 95 sites) by machine learning. The calculation of damage was the difference between the dry matter yield (DMY)_normal and DMY_abnormal. The normal climate was set as the 40-year of climate data according to the year of WCC data (1978-2017). The level of abnormal climate by temperature and precipitation was set as RCP 4.5 standard. The DMY_normal ranged from 13,845-19,347 kg/ha. The damage of WCC which was differed depending on the region and level of abnormal climate where abnormal temperature and precipitation occurred. The damage of abnormal temperature in 2050 and 2100 ranged from -263 to 360 and -1,023 to 92 kg/ha, respectively. The damage of abnormal precipitation in 2050 and 2100 was ranged from -17 to 2 and -12 to 2 kg/ha, respectively. The maximum damage was 360 kg/ha that the abnormal temperature in 2050. As the average monthly temperature increases, the DMY of WCC tends to increase. The damage calculated through the RCP 4.5 standard was presented as a mapping using QGIS. Although this study applied the scenario in which greenhouse gas reduction was carried out, additional research needs to be conducted applying an RCP scenario in which greenhouse gas reduction is not performed.

• Journal of Digital Convergence
• /
• v.17 no.8
• /
• pp.249-255
• /
• 2019

Type 2 diabetes mellitus(T2DM) is included in metabolic disorders characterized by hyperglycemia, which causes many complications, and requires long-term treatment resulting in massive medical expenses each year. There have been many studies to solve this problem, but the existing studies have not been accurate by learning and predicting the data at specific time point. Thus, this study proposed a model using RNN to increase the accuracy of prediction of T2DM. This work propose a T2DM prediction model based on Korean Genome and Epidemiology study(Ansan, Anseong Korea). We trained all of the data over time to create prediction model of diabetes. To verify the results of the prediction model, we compared the accuracy with the existing machine learning methods, LR, k-NN, and SVM. Proposed prediction model accuracy was 0.92 and the AUC was 0.92, which were higher than the other. Therefore predicting the onset of T2DM by using the proposed diabetes prediction model in this study, it could lead to healthier lifestyle and hyperglycemic control resulting in lower risk of diabetes by alerted diabetes occurrence.

• Journal of IKEEE
• /
• v.20 no.4
• /
• pp.351-360
• /
• 2016

Recently, machine learning algorithms, especially deep learning-based algorithms, have been receiving attention due to its high classification performance. Among the algorithms, Convolutional Neural Network(CNN) is known to be efficient for image processing tasks used for Advanced Driver Assistance Systems(ADAS). However, it is difficult to achieve real-time processing for CNN in vehicle embedded software environment due to the repeated operations contained in each layer of CNN. In this paper, we propose a hardware accelerator which enhances the execution time of CNN by parallelizing the repeated operations such as convolution. Xilinx ZC706 evaluation board is used to verify the performance of the proposed accelerator. For $36{\times}36$ input images, the hardware execution time of CNN is 2.812ms in 100MHz clock frequency and shows that our hardware can be executed in real-time.

• Journal of Advanced Navigation Technology
• /
• v.22 no.5
• /
• pp.478-484
• /
• 2018

This study predicts solar radiation, solar radiation, and solar power generation using hourly weather data such as temperature, precipitation, wind direction, wind speed, humidity, cloudiness, sunshine and solar radiation. I/O pattern in supervised learning is the most important factor in prediction, but it must be determined by repeated experiments because humans have to decide. This study proposed four input and output patterns for solar and sunrise prediction. In addition, we predicted solar power generation using the predicted solar and solar radiation data and power generation data of Youngam solar power plant in Jeollanamdo. As a experiment result, the model 4 showed the best prediction results in the sunshine and solar radiation prediction, and the RMSE of sunshine was 1.5 times and the sunshine RMSE was 3 times less than that of model 1. As a experiment result of solar power generation prediction, the best prediction result was obtained for model 4 as well as sunshine and solar radiation, and the RMSE was reduced by 2.7 times less than that of model 1.

• Journal of Digital Convergence
• /
• v.19 no.9
• /
• pp.257-269
• /
• 2021

The purpose of this study is to collect biosignal data in a non-invasive and non-restrictive manner using a BCG (Ballistocardiogram) sensor, and utilize artificial intelligence machine learning algorithms in ICT and high-performance computing environments. And it is to present and study a method for developing and validating a data-based blood glucose prediction model. In the blood glucose level prediction model, the input nodes in the MLP architecture are data of heart rate, respiration rate, stroke volume, heart rate variability, SDNN, RMSSD, PNN50, age, and gender, and the hidden layer 7 were used. As a result of the experiment, the average MSE, MAE, and RMSE values of the learning data tested 5 times were 0.5226, 0.6328, and 0.7692, respectively, and the average values of the validation data were 0.5408, 0.6776, and 0.7968, respectively, and the coefficient of determination (R²) was 0.9997. If research to standardize a model for predicting blood sugar levels based on data and to verify data set collection and prediction accuracy continues, it is expected that it can be used for non-invasive blood sugar level management.

• Journal of Digital Convergence
• /
• v.17 no.11
• /
• pp.233-240
• /
• 2019

In order to predict and respond to cyber attacks, a number of security companies quickly identify the methods, types and characteristics of attack techniques and are publishing Security Intelligence Reports(SIRs) on them. However, the SIRs distributed by each company are huge and unstructured. In this paper, we propose a framework that uses five analytic techniques to formulate a report and extract key information in order to reduce the time required to extract information on large unstructured SIRs efficiently. Since the SIRs data do not have the correct answer label, we propose four analysis techniques, Keyword Extraction, Topic Modeling, Summarization, and Document Similarity, through Unsupervised Learning. Finally, has built the data to extract threat information from SIRs, analysis applies to the Named Entity Recognition (NER) technology to recognize the words belonging to the IP, Domain/URL, Hash, Malware and determine if the word belongs to which type We propose a framework that applies a total of five analysis techniques, including technology.