• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 추계학술발표대회
• /
• pp.942-943
• /
• 2021

The present work proposes a classification system for the HEp-2 cell images using an unsupervised deep feature learning method. Unlike most of the state-of-the-art methods in the literature that utilize deep learning in a strictly supervised way, we propose here the use of the deep convolutional autoencoder (DCAE) as the principal feature extractor for classifying the different types of the HEp-2 cell images. The network takes the original cell images as the inputs and learns to reconstruct them in order to capture the features related to the global shape of the cells. A final feature vector is constructed by using the latent representations extracted from the DCAE, giving a highly discriminative feature representation. The created features will be fed to a nonlinear classifier whose output will represent the final type of the cell image. We have tested the discriminability of the proposed features on one of the most popular HEp-2 cell classification datasets, the SNPHEp-2 dataset and the results show that the proposed features manage to capture the distinctive characteristics of the different cell types while performing at least as well as the actual deep learning based state-of-the-art methods.

• 정보보호학회논문지
• /
• 제30권4호
• /
• pp.617-629
• /
• 2020

인터넷 컴퓨팅 환경의 변화, 새로운 서비스 출현, 그리고 지능화되어 가는 해커들의 다양한 공격으로 인한 규칙 기반 침입탐지시스템의 한계점을 극복하기 위해 기계학습 및 딥러닝 기술을 활용한 네트워크 이상 검출(NAD: Network Anomaly Detection)에 대한 관심이 집중되고 있다. NAD를 위한 대부분의 기존 기계학습 및 딥러닝 기술은 '정상'과 '공격'으로 레이블링된 훈련용 데이터 셋을 학습하는 지도학습 방법을 사용한다. 본 논문에서는 공격의 징후가 없는 일상의 네트워크에서 수집할 수 있는 레이블링이 필요 없는 데이터 셋을 이용하는 비지도학습 오토 엔코더(AE: AutoEncoder)를 활용한 NAD 적용 가능성을 제시한다. AE 성능을 검증하기 위해 NSL-KDD 훈련 및 시험 데이터 셋을 사용해 정확도, 정밀도, 재현율, f1-점수, 그리고 ROC AUC (Receiver Operating Characteristic Area Under Curve) 값을 보인다. 특히 이들 성능지표를 대상으로 AE의 층수, 규제 강도, 그리고 디노이징 효과 등을 분석하여 레퍼런스 모델을 제시하였다. AE의 훈련 데이터 셋에 대한 재생오류 82-th 백분위수를 기준 값으로 KDDTest+와 KDDTest-21 시험 데이터 셋에 대해 90.4%와 89% f1-점수를 각각 보였다.

• 대한원격탐사학회지
• /
• 제39권1호
• /
• pp.1-21
• /
• 2023

In this technically advanced era, we are surrounded by smartphones, computers, and cameras, which help us to store visual information in 2D image planes. However, such images lack 3D spatial information about the scene, which is very useful for scientists, surveyors, engineers, and even robots. To tackle such problems, depth maps are generated for respective image planes. Depth maps or depth images are single image metric which carries the information in three-dimensional axes, i.e., xyz coordinates, where z is the object's distance from camera axes. For many applications, including augmented reality, object tracking, segmentation, scene reconstruction, distance measurement, autonomous navigation, and autonomous driving, depth estimation is a fundamental task. Much of the work has been done to calculate depth maps. We reviewed the status of depth map estimation using different techniques from several papers, study areas, and models applied over the last 20 years. We surveyed different depth-mapping techniques based on traditional ways and newly developed deep-learning methods. The primary purpose of this study is to present a detailed review of the state-of-the-art traditional depth mapping techniques and recent deep learning methodologies. This study encompasses the critical points of each method from different perspectives, like datasets, procedures performed, types of algorithms, loss functions, and well-known evaluation metrics. Similarly, this paper also discusses the subdomains in each method, like supervised, unsupervised, and semi-supervised methods. We also elaborate on the challenges of different methods. At the conclusion of this study, we discussed new ideas for future research and studies in depth map research.

• 전자공학회논문지CI
• /
• 제38권1호
• /
• pp.1-6
• /
• 2001

본 논문에서는 Hebb 학습법에 기초한 Kohonen의 LVQ 학습법을 퍼셉트론 학습에 사용되는 경도 강하 (Gradient descent) 학습법에 의해 재해석한다. Kohonen의 LVQ는 학습법에 따라 두 가지로 나뉠 수 있는데 하나는 자율학습 LVQ(ULVQ)이며 다른 하나는 타율학습 LVQ(SLVQ)이다. 두 경우 모두 출력뉴런의 목표 값을 적당히 생성할 경우 타율학습 경도 강하학습법으로 표현될 수 있다. 결과적으로 LVQ학습법은 타율학습 경도 강하 학습법의 특수한 형태임을 알 수 있으며 또한 LVQ는 보다 일반화된 '퍼셉트론 형태의 LVQ(PLVQ)'알고리즘으로 표현될 수 있음을 알 수 있다. 본 논문에서는 이를 증명하고 결론을 맺는다.

• International Journal of Contents
• /
• 제16권2호
• /
• pp.8-18
• /
• 2020

Data is critical in deep learning but the scarcity of data often occurs in research, especially in the preparation of the paired training data. In this paper, document image binarization with unpaired data is studied by introducing adversarial learning, excluding the need for supervised or labeled datasets. However, the simple extension of the previous unpaired training to binarization inevitably leads to poor performance compared to paired data training. Thus, a new deep learning approach is proposed by introducing a multi-diversity of higher quality generated images. In this paper, a two-stage model is proposed that comprises the generative adversarial network (GAN) followed by the U-net network. In the first stage, the GAN uses the unpaired image data to create paired image data. With the second stage, the generated paired image data are passed through the U-net network for binarization. Thus, the trained U-net becomes the binarization model during the testing. The proposed model has been evaluated over the publicly available DIBCO dataset and it outperforms other techniques on unpaired training data. The paper shows the potential of using unpaired data for binarization, for the first time in the literature, which can be further improved to replace paired data training for binarization in the future.

• 품질경영학회지
• /
• 제48권3호
• /
• pp.421-431
• /
• 2020

Purpose: The purpose of this study is to compare and analyze the tokenizer in natural language processing for customer satisfaction in sentiment analysis. Methods: In this study, a supervised learning-based tokenizer Mecab-Ko and an unsupervised learning-based tokenizer SentencePiece were used for comparison. Three algorithms: Naïve Bayes, k-Nearest Neighbor, and Decision Tree were selected to compare the performance of each tokenizer. For performance comparison, three metrics: accuracy, precision, and recall were used in the study. Results: The results of this study are as follows; Through performance evaluation and verification, it was confirmed that SentencePiece shows better classification performance than Mecab-Ko. In order to confirm the robustness of the derived results, independent t-tests were conducted on the evaluation results for the two types of the tokenizer. As a result of the study, it was confirmed that the classification performance of the SentencePiece tokenizer was high in the k-Nearest Neighbor and Decision Tree algorithms. In addition, the Decision Tree showed slightly higher accuracy among the three classification algorithms. Conclusion: The SentencePiece tokenizer can be used to classify and interpret customer sentiment based on online reviews in Korean more accurately. In addition, it seems that it is possible to give a specific meaning to a short word or a jargon, which is often used by users when evaluating products but is not defined in advance.

• 지능정보연구
• /
• 제15권4호
• /
• pp.23-36
• /
• 2009

네트워크 침입 탐지는 데이터마이닝 기법을 활용하면서 지속적으로 발전하여 왔다. 데이터마이닝에 의한 침입 탐지 기법에는 클래스 레이블을 이용한 감독 학습과 클래스 레이블이 없는 비감독 학습 방법이 있다. 본 논문에서는 클래스 레이블이 없는 비감독 학습 방법인 LBG 클러스터링 알고리즘을 이용하여 네트워크 침입 탐지 정확도를 높이는 방법을 연구하였다. 임의의 초기 중심값들로 시작하여 유클리디언 거리 기반에 의해 클러스터링을 수행하는 K-means 방법은 잡음(noisy) 데이터와 이상치(outlier)에 대하여 취약하다는 단점이 있다. 비균일이진 분할에 의한 클러스터링 알고리즘은 초기값 없이 이진분할에 의해 클러스터링을 수행하며 수행 속도가 빠르다. 본 논문에서는 이 두 알고리즘의 장단점을 통합한 EM(Expectation Maximization) 기반의 LBG 알고리즘을 네트워크 침입 탐지에 적용하였으며, KDD 컵 데이터셋을 대상으로 한 실험을 통하여 LBG 알고리즘을 이용함으로써 침입 탐지의 정확도를 높일 수 있음을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권6호
• /
• pp.2470-2491
• /
• 2018

Web-scale open information extraction (Open IE) plays an important role in NLP tasks like acquiring common-sense knowledge, learning selectional preferences and automatic text understanding. A large number of Open IE approaches have been proposed in the last decade, and the majority of these approaches are based on supervised learning or dependency parsing. In this paper, we present a novel method for web scale open information extraction, which employs cosine distance based on Google word vector as the confidence score of the extraction. The proposed method is a purely unsupervised learning algorithm without requiring any hand-labeled training data or dependency parse features. We also present the mathematically rigorous proof for the new method with Bayes Inference and Artificial Neural Network theory. It turns out that the proposed algorithm is equivalent to Maximum Likelihood Estimation of the joint probability distribution over the elements of the candidate extraction. The proof itself also theoretically suggests a typical usage of word vector for other NLP tasks. Experiments show that the distance-based method leads to further improvements over the newly presented Open IE systems on three benchmark datasets, in terms of effectiveness and efficiency.

• 한국지능시스템학회논문지
• /
• 제8권2호
• /
• pp.106-116
• /
• 1998

유연 로봇 매니퓰레이터의 위치 제어 알고리즘에 대한 연구를 하였다. 제안하는 알고리즘은 신경회로망의 학습 알고리즘에 근거한 자동 구축 퍼지 적응 제어기(ACFAC : Automaitc Constructed Fuzzy Adaptive controller)에 기본으로 한다. 제안하는 시스템은 비지도 경쟁 학습 알고리즘을 사용하여 입력 변수의 멤버십 함수와 지도 Outstar 학습 알고리즘을 사용하여 출력 정보를 학습시킨다. ACFAC는 유연 로봇 매니퓰레이터의 동력한 모델을 필요로 하지 않는다. ACFAC는 유연 로봇 매니퓰레이터의 끝점이 원하는 궤적을 따라가도록 설계되었다. 이 제어기의 입력은 위치 오차, 위치 오차의 미분 값과 오차의 variation에 의해 결정된다. ACFAC의 우수서을 보여주기 우해서 PID 제어나 신경회로망 알고리즘을 사용한 결과와 비교를 하였다.

• 산업경영시스템학회지
• /
• 제43권1호
• /
• pp.123-131
• /
• 2020

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.