• 한국빅데이터학회지
• /
• 제5권1호
• /
• pp.1-9
• /
• 2020

이 연구는 2018년도 정부(농림축산식품부)의 재원으로 농림식품기술기획평가원 지원을 받아 수행된 연구이다. 최근 시계열 및 텍스트 마이닝에서 활발히 사용되는 모델은 딥러닝(Deep Learning) 모델 구조를 활용한 LSTM(Long Short-Term Memory models) 모델이다. LSTM 모델은 RNN의 BPTT(Backpropagation Through Time) 과정에서 발생하는 Long-Term Dependency Problem을 해결하기 위해 등장한 모델이다. LSTM 모델은 가변적인 Sequence data를 활용하여 예측하는 문제를 굉장히 잘 해결했고, 지금도 널리 사용되고 있다. 본 논문 연구에서는 KT가 제공하는 CDR(Call Detailed Record) 데이터를 활용하여 바이러스와 밀접한 관계가 있을 것으로 예측되는 사람의 이동 경로를 파악하였다. 해당 사람의 경로를 활용하여 LSTM 모델을 학습시켜 이동 경로를 예측한 결과를 소개한다. 본 연구 결과를 활용하여 HPAI가 전파되는 경로를 예측하여 방역에 중점을 둘 경로 또는 지역을 선정해 HPAI 확산을 줄이는 데 이용될 수 있을 것이다.

• 대한의용생체공학회:의공학회지
• /
• 제40권6호
• /
• pp.242-249
• /
• 2019

Forearm electromyography (EMG) generated by wrist movements has been widely used to develop an electrical prosthetic hand, but EMG generated by finger movements has been rarely used even though 20% of amputees lose fingers. The goal of this study is to improve the classification performance of different finger movements using a deep learning algorithm, and thereby contributing to the development of a high-performance finger-based prosthetic hand. Ten participants took part in this study, and they performed seven different finger movements forty times each (thumb, index, middle, ring, little, fist and rest) during which EMG was measured from the back of the right hand using four bipolar electrodes. We extracted mean absolute value (MAV), root mean square (RMS), and mean (MEAN) from the measured EMGs for each trial as features, and a 5x5-fold cross-validation was performed to estimate the classification performance of seven different finger movements. A long short-term memory (LSTM) model was used as a classifier, and linear discriminant analysis (LDA) that is a widely used classifier in previous studies was also used for comparison. The best performance of the LSTM model (sensitivity: 91.46 ± 6.72%; specificity: 91.27 ± 4.18%; accuracy: 91.26 ± 4.09%) significantly outperformed that of LDA (sensitivity: 84.55 ± 9.61%; specificity: 84.02 ± 6.00%; accuracy: 84.00 ± 5.87%). Our result demonstrates the feasibility of a deep learning algorithm (LSTM) to improve the performance of classifying different finger movements using EMG.

• 대한임베디드공학회논문지
• /
• 제14권5호
• /
• pp.277-286
• /
• 2019

In this paper, we propose an efficient hyperplane generation technique to classify human activity from combination of events and sequence information obtained from multiple-event sensors. By generating hyperplane efficiently, our machine learning algorithm classify with less memory and run time than the LSVM (Linear Support Vector Machine) for embedded system. Because the fact that light weight and high speed algorithm is one of the most critical issue in the IoT, the study can be applied to smart home to predict human activity and provide related services. Our approach is based on reducing numbers of hyperplanes and utilizing robust string comparing algorithm. The proposed method results in reduction of memory consumption compared to the conventional ML (Machine Learning) algorithms; 252 times to LSVM and 34,033 times to LSTM (Long Short-Term Memory), although accuracy is decreased slightly. Thus our method showed outstanding performance on accuracy per hyperplane; 240 times to LSVM and 30,520 times to LSTM. The binarized image is then divided into groups, where each groups are converted to binary number, in order to reduce the number of comparison done in runtime process. The binary numbers are then converted to string. The test data is evaluated by converting to string and measuring similarity between hyperplanes using Levenshtein algorithm, which is a robust dynamic string comparing algorithm. This technique reduces runtime and enables the proposed algorithm to become 27% faster than LSVM, and 90% faster than LSTM.

• 자동차안전학회지
• /
• 제13권4호
• /
• pp.129-143
• /
• 2021

This paper proposes actuator fault detection and adaptive fault-tolerant control algorithms using performance index and human-like learning for longitudinal autonomous vehicles. Conventional longitudinal controller for autonomous driving consists of supervisory, upper level and lower level controllers. In this paper, feedback control law and PID control algorithm have been used for upper level and lower level controllers, respectively. For actuator fault-tolerant control, adaptive rule has been designed using the gradient descent method with estimated coefficients. In order to adjust the control parameter used for determination of adaptation gain, human-like learning algorithm has been designed based on perceptron learning method using control errors and control parameter. It is designed that the learning algorithm determines current control parameter by saving it in memory and updating based on the cost function-based gradient descent method. Based on the updated control parameter, the longitudinal acceleration has been computed adaptively using feedback law for actuator fault-tolerant control. The finite window-based performance index has been designed for detection and evaluation of actuator performance degradation using control error.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2188-2203
• /
• 2021

With the rapid development of mobile Internet, smart phones have been widely popularized, among which Android platform dominates. Due to it is open source, malware on the Android platform is rampant. In order to improve the efficiency of malware detection, this paper proposes deep learning Android malicious detection system based on behavior features. First of all, the detection system adopts the static analysis method to extract different types of behavior features from Android applications, and extract sensitive behavior features through Term frequency-inverse Document Frequency algorithm for each extracted behavior feature to construct detection features through unified abstract expression. Secondly, Long Short-Term Memory neural network model is established to select and learn from the extracted attributes and the learned attributes are used to detect Android malicious applications, Analysis and further optimization of the application behavior parameters, so as to build a deep learning Android malicious detection method based on feature analysis. We use different types of features to evaluate our method and compare it with various machine learning-based methods. Study shows that it outperforms most existing machine learning based approaches and detects 95.31% of the malware.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권4호
• /
• pp.478-487
• /
• 2004

메모리 기반 추론에서 기억공간의 효율적인 사용과 분류성능의 향상을 위하여 제안되었던 RPA(Recursive Partition Averaging)알고리즘은 대상 패턴 공간을 분할 한 후 대표 패턴을 추출하여 분류 기준 패턴으로 사용한다. 이 기법은 메모리 사용 효율과 분류 성능 면에서 우수한 결과를 보였지만, 분할 종료 조건과 대표패턴의 추출 방법이 분류 성능 저하의 원인이 되는 단점을 가지고 있었다. 여기에서는 기존 RPA의 단점을 보안한 ARPA(Adaptive RPA) 알고리즘을 제안한다. 제안된 알고리즘은 패턴 공간의 분할 종료 조건으로 특징별 최빈 패턴 구간(FPD: Feature-based population densimeter)추출 알고리즘을 사용하며, 학습 결과 패턴의 생성을 대표패턴 추출기법 대신 최빈 패턴 구간을 이용하여 생성한 최적초월평면(OH： Optimized Hyperrectangle)을 사용한다. 제안된 알고리즘은 k-NN 분류기에서 필요로 하는 메모리 공간의 40%정도를 사용하며, 분류에 있어서도 RPA보다 우수한 인식 성능을 보이고 있다. 또한 저장된 패턴의 감소로 인하여, 실제 분류에 소요되는 시간 비교에 있어서도 k-NN보다 월등히 우수한 성능을 보이고 있다.

• 수산해양교육연구
• /
• 제16권2호
• /
• pp.163-170
• /
• 2004

The rapid exchanges of mobile computing environment and development of wireless communication are providing many effects for learning activity of students. Recently, PDA system developers which are studying memory capacity, communication speed and size of screen support techniques to be capable of learning from students in the wireless or moving environment. In this viewpoints, this paper has a purpose to design multimedia learning system to be able to do with sound lecture contents. The implemented system largely consists of two parts which have the teacher module and students module. The one manages learning progress of students, class management, bulletin board and etc. The other is capable of using studying and bulletin functions. The main idea of this research is focus to upgrade the effect of learning without almost treating the existing studies, which can be listening sound lecture and also seeing text and image at the same time.

• International Journal of Computer Science & Network Security
• /
• 제22권2호
• /
• pp.131-138
• /
• 2022

Soft sensors are used to anticipate complicated model parameters using data from classifiers that are comparatively easy to gather. The goal of this study is to use artificial intelligence techniques to design and build soft sensors. The combination of a Long Short-Term Memory (LSTM) network and Grey Wolf Optimization (GWO) is used to create a unique soft sensor. LSTM is developed to tackle linear model with strong nonlinearity and unpredictability of manufacturing applications in the learning approach. GWO is used to accomplish input optimization technique for LSTM in order to reduce the model's inappropriate complication. The newly designed soft sensor originally brought LSTM's superior dynamic modeling with GWO's exact variable selection. The performance of our proposal is demonstrated using simulations on real-world datasets.

• Journal of Information Processing Systems
• /
• 제18권1호
• /
• pp.12-25
• /
• 2022

In recent years, deep convolutional neural networks have made significant progress in the research of single image super-resolution. However, it is difficult to be applied in practical computing terminals or embedded devices due to a large number of parameters and computational effort. To balance these problems, we propose CSRNet, a lightweight neural network based on channel split residual learning structure, to reconstruct highresolution images from low-resolution images. Lightweight refers to designing a neural network with fewer parameters and a simplified structure for lower memory consumption and faster inference speed. At the same time, it is ensured that the performance of recovering high-resolution images is not degraded. In CSRNet, we reduce the parameters and computation by channel split residual learning. Simultaneously, we propose a double-upsampling network structure to improve the performance of the lightweight super-resolution network and make it easy to train. Finally, we propose a new evaluation metric for the lightweight approaches named 100_FPS. Experiments show that our proposed CSRNet not only speeds up the inference of the neural network and reduces memory consumption, but also performs well on single image super-resolution.

• ETRI Journal
• /
• 제46권3호
• /
• pp.379-391
• /
• 2024

To meet increasing traffic requirements in mobile networks, small base stations (SBSs) are densely deployed, overlapping existing network architecture and increasing system capacity. However, densely deployed SBSs increase energy consumption and interference. Although these problems already exist because of densely deployed SBSs, even more SBSs are needed to meet increasing traffic demands. Hence, base station (BS) switching operations have been used to minimize energy consumption while guaranteeing quality-of-service (QoS) for users. In this study, to optimize energy efficiency, we propose the use of deep reinforcement learning (DRL) to create a BS switching operation strategy with a traffic prediction model. First, a federated long short-term memory (LSTM) model is introduced to predict user traffic demands from user trajectory information. Next, the DRL-based BS switching operation scheme determines the switching operations for the SBSs using the predicted traffic demand. Experimental results confirm that the proposed scheme outperforms existing approaches in terms of energy efficiency, signal-to-interference noise ratio, handover metrics, and prediction performance.