• Smart Media Journal
• /
• v.9 no.1
• /
• pp.38-44
• /
• 2020

Due to the high dependence on electric energy, electric fires make up a significant portion of fires in Korea. Electric arcs by short circuits or poor contact cause three of four electrical fires. An electric arc is a discharge phenomenon of electrical current between the insulators, which instantaneously produces high temperature. In order to reduce the fire due to electric arc, this study aims to predict the electric arc risk. We collected arc data from the arc detectors and converted into graphs based on temporal arc data. We used machine learning for training converted graph with different number of temporal arc data. To measure the performance of the learning model, we use the test data. In the results, when the number of temporal arc data was 20, the prediction rate was high as 86%.

• Annual Conference of KIPS
• /
• 2020.11a
• /
• pp.349-352
• /
• 2020

클라우드는 가상화 기술을 사용한 리소스의 유연성과 뛰아난 접근성을 장점으로 빅데이터, 딥러닝 등 여러 분야에서 클라우드를 사용하고 있다. 최근 클라우드와 결합된 IoT 시스템을 통해 시스템 관리, 데이터 처리 및 저장, 데이터를 이용한 빅데이터 활용 등 여러 방법으로 사용 할 수 있어 많은 관심을 받고 있다. 그러나 IoT 클라우드의 많은 활용에 따라 대규모 시스템화, 여러 사용자의 개인정보 저장 등의 이유로 많은 공격자의 표적이 되고있다. 여러 공격자의 공격을 방아하기 위해 IoT 클라우드 시스템은 블록체인, 보안 IoT 디바이스, 변형된 클라우드 모델등 여러 연구가 진행되고 있다. 본 논문에서는 최근 연구되고 있는 블록체인, 클라우드, IoT 시스템의 동향에 대해 조사하고, 기존에 연구되었던 기술을 바탕으로 효과적인 블록체인 기반의 IoT 클라우드 시스템을 제안한다. 제안하는 IoT 클라우드 시스템은 블록체인 기술을 사용하여 보안정책을 관리할 수 있어 신뢰성이 높으며, 클라우드 시스템이 작동하지 않을 경우 페일오버 기능을 수행할 수 있어 가용성이 뛰어나다.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.19 no.6
• /
• pp.915-936
• /
• 2017

In this study, current status of Korean hazard mitigation guideline for tunnel operation is summarized. It shows that requirement for CCTV installation has been gradually stricted and needs for tunnel incident detection system in conjunction with the CCTV in tunnels have been highly increased. Despite of this, it is noticed that mathematical algorithm based incident detection system, which are commonly applied in current tunnel operation, show very low detectable rates by less than 50%. The putative major reasons seem to be (1) very weak intensity of illumination (2) dust in tunnel (3) low installation height of CCTV to about 3.5 m, etc. Therefore, an attempt in this study is made to develop an deep-learning based tunnel incident detection system, which is relatively insensitive to very poor visibility conditions. Its theoretical background is given and validating investigation are undertaken focused on the moving vehicles and person out of vehicle in tunnel, which are the official major objects to be detected. Two scenarios are set up: (1) training and prediction in the same tunnel (2) training in a tunnel and prediction in the other tunnel. From the both cases, targeted object detection in prediction mode are achieved to detectable rate to higher than 80% in case of similar time period between training and prediction but it shows a bit low detectable rate to 40% when the prediction times are far from the training time without further training taking place. However, it is believed that the AI based system would be enhanced in its predictability automatically as further training are followed with accumulated CCTV BigData without any revision or calibration of the incident detection system.

• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.822-828
• /
• 2018

Recently, energy issues such as massive blackout due to increase in power consumption have been emerged, and it is necessary to improve the accuracy of prediction of power consumption as a solution for these problems. In this study, we investigate the difference between the actual power consumption and the predicted power consumption through the deep learning- based power consumption forecasting experiment, and the possibility of adjusting the power reserve ratio. In this paper, the prediction of the power consumption based on the deep learning can be used as a basis to reduce the power reserve ratio so as not to excessively produce extra power. The deep learning method used in this paper uses a learning model of long-short-term-memory (LSTM) structure that processes time series data. In the computer simulation, the generated power consumption data was learned, and the power consumption was predicted based on the learned model. We calculate the error between the actual and predicted power consumption amount, resulting in an error rate of 21.37%. Considering the recent power reserve ratio of 45.9%, it is possible to reduce the reserve ratio by 20% when applying the power consumption prediction algorithm proposed in this study.

• Journal of IKEEE
• /
• v.22 no.3
• /
• pp.829-835
• /
• 2018

Recently, studies have been actively conducted to reduce spare power that is unnecessarily generated or wasted in existing power systems and to improve energy use efficiency. In this study, smart meter, which is one of the element technologies of smart grid, is implemented to improve the efficiency of energy use by controlling power of electric devices, and predicting trends of energy usage based on deep learning. We propose and develop an algorithm that controls the power of the electric devices by comparing the predicted power consumption with the real-time power consumption. To verify the performance of the proposed smart meter based on the deep running, we constructed the actual power consumption environment and obtained the power usage data in real time, and predicted the power consumption based on the deep learning model. We confirmed that the unnecessary power consumption can be reduced and the energy use efficiency increases through the proposed deep learning-based smart meter.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.101-125
• /
• 2022

Recently, with the development of computing technology and the improvement of the cloud environment, deep learning technology has developed, and attempts to apply deep learning to various fields are increasing. A typical example is anomaly detection, which is a technique for identifying values or patterns that deviate from normal data. Among the representative types of anomaly detection, it is very difficult to detect a contextual anomaly that requires understanding of the overall situation. In general, detection of anomalies in image data is performed using a pre-trained model trained on large data. However, since this pre-trained model was created by focusing on object classification of images, there is a limit to be applied to anomaly detection that needs to understand complex situations created by various objects. Therefore, in this study, we newly propose a two-step pre-trained model for detecting abnormal situation. Our methodology performs additional learning from image captioning to understand not only mere objects but also the complicated situation created by them. Specifically, the proposed methodology transfers knowledge of the pre-trained model that has learned object classification with ImageNet data to the image captioning model, and uses the caption that describes the situation represented by the image. Afterwards, the weight obtained by learning the situational characteristics through images and captions is extracted and fine-tuning is performed to generate an anomaly detection model. To evaluate the performance of the proposed methodology, an anomaly detection experiment was performed on 400 situational images and the experimental results showed that the proposed methodology was superior in terms of anomaly detection accuracy and F1-score compared to the existing traditional pre-trained model.

• Journal of KIISE
• /
• v.44 no.8
• /
• pp.803-812
• /
• 2017

T-commerce is technology-fusion service on which the user can purchase using data broadcasting technology based on bi-directional digital TVs. To achieve the best revenue under a limited environment in regard to the channel number and the variety of sales goods, organizing broadcast programs to maximize the expected sales considering the selling power of each product at each time slot. For this, this paper proposes a method to predict the sales of goods when it is assigned to each time slot. The proposed method predicts the sales of product at a time slot given the week-in-year and weather of the target day. Additionally, it combines a statistical predict model applying SVD (Singular Value Decomposition) to mitigate the sparsity problem caused by the bias in sales record. In experiments on the sales data of W-shopping, a T-commerce company, the proposed method showed NMAE (Normalized Mean Absolute Error) of 0.12 between the prediction and the actual sales, which confirms the effectiveness of the proposed method. The proposed method is practically applied to the T-commerce system of W-shopping and used for broadcasting organization.

• Journal of Digital Convergence
• /
• v.18 no.11
• /
• pp.259-266
• /
• 2020

Many developments have been steadily carried out by researchers with applying knowledge-based expert system or machine learning algorithms to the financial field. In particular, it is now common to perform knowledge based system trading in using stock prices. Recently, deep learning technologies have been applied to real fields of stock trading marketplace as GPU performance and large scaled data have been supported enough. Especially, LSTM has been tried to apply to stock price prediction because of its compatibility for time series data. In this paper, we implement stock price prediction using LSTM. In modeling of LSTM, we propose a fitness combination of model parameters and activation functions for best performance. Specifically, we propose suitable selection methods of initializers of weights and bias, regularizers to avoid over-fitting, activation functions and optimization methods. We also compare model performances according to the different selections of the above important modeling considering factors on the real-world stock price data of global major companies. Finally, our experimental work brings a fitness method of applying LSTM model to stock price prediction.

• 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.

• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.111-115
• /
• 2017

목적 지향적 대화 시스템(Goal-oriented dialogue system) 은 텍스트나 음성을 통해 특정한 목적을 수행 할 수 있는 시스템이다. 최근 RNN(recurrent neural networks)을 기반으로 대화 데이터를 end-to-end learning 방식으로 학습하여 대화 시스템을 구축하는데에 활용한 연구가 있다. End-to-end 방식의 학습은 도메인에 대한 지식 없이 학습 데이터 자체만으로 대화 시스템 구축을 위한 학습이 가능하다는 장점이 있지만 도메인 지식을 학습하기 위해서는 많은 양의 데이터가 필요하다는 단점이 존재한다. 이에 본 논문에서는 도메인 특정 지식을 결합하여 end-to-end learning 방식의 학습이 가능한 Hybrid Code Network 구조를 기반으로 한국어로 구성된 식당 예약에 관련한 대화 데이터셋을 이용하여 식당 예약을 목적으로하는 대화 시스템을 구축하는 방법을 제안한다. 실험 결과 본 시스템은 응답 별 정확도 95%와 대화 별 정확도 63%의 성능을 나타냈다.