• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.480-482
• /
• 2012

하이퍼네트워크(Hypernetworks)는 하이퍼에지(hyperedge)들로 이루어진 생성 모델(generative model)로서, 주로 이산(binary) 데이터에 적용되어왔다. 본 논문에서는 이산 데이터와 실수 데이터를 모두 다룰 수 있는 새로운 하이퍼네트워크 모델을 에너지 기반 모델(energy-based model)의 형태로 제시하고, 비지도 학습(unsupervised learning) 알고리즘으로 데이터를 성공적으로 학습함을 간단한 실험을 통해 보이겠다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.111-116
• /
• 2016

의미역은 자연어 문장의 서술어와 관련된 논항의 역할을 설명하는 것으로, 주어진 서술어에 대한 논항 인식(Argument Identification) 및 분류(Argument Labeling)의 과정을 거쳐 의미역 결정(Semantic Role Labeling)이 이루어진다. 이를 위해서는 격틀 사전을 이용한 방법이나 말뭉치를 이용한 지도 학습(Supervised Learning) 방법이 주를 이루고 있다. 이때, 격틀 사전 또는 의미역 주석 정보가 부착된 말뭉치를 구축하는 것은 필수적이지만, 이러한 노력을 최소화하기 위해 본 논문에서는 비모수적 베이지안 모델(Nonparametric Bayesian Model)을 기반으로 서술어에 가능한 의미역을 추론하는 비지도 학습(Unsupervised Learning)을 수행한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.3
• /
• pp.129-134
• /
• 2015

Since automatic social engineering based spam attacks induce for users to click or receive the short message service (SMS), e-mail, site address and make a relationship with an unknown friend, it is very easy for them to active in online social networks. The previous spam detection schemes only apply manual filtering of the system managers or labeling classifications regardless of the features of social networks. In this paper, we propose the spam detection metric after reflecting on a couple of features of social networks followed by analysis of real social network data set, Twitter spam. In addition, we provide the online social networks based unsupervised scheme for automated social engineering spam with self organizing map (SOM). Through the performance evaluation, we show the detection accuracy up to 90% and the possibility of real time training for the spam detection without the manager.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.11
• /
• pp.41-49
• /
• 2019

We are to build an unsupervised machine learning-based language model which can estimate the amount of information that are in need to process words consisting of subword-level morphemes and syllables. We are then to investigate whether the reading times of words reflecting their morphemic and syllabic structures are predicted by an information-theoretic measure such as surprisal. Specifically, the proposed Morfessor-based unsupervised machine learning model is first to be trained on the large dataset of sentences on Sejong Corpus and is then to be applied to estimate the information-theoretic measure on each word in the test data of Korean words. The reading times of the words in the test data are to be recruited from Korean Lexicon Project (KLP) Database. A comparison between the information-theoretic measures of the words in point and the corresponding reading times by using a linear mixed effect model reveals a reliable correlation between surprisal and reading time. We conclude that surprisal is positively related to the processing effort (i.e. reading time), confirming the surprisal hypothesis.

• Journal of Convergence for Information Technology
• /
• v.12 no.4
• /
• pp.260-269
• /
• 2022

In this paper, We propose an anomaly detection model using deep neural network to automate the identification of outliers of the national air pollution measurement network data that is previously performed by experts. We generated training data by analyzing missing values and outliers of weather data provided by the Institute of Environmental Research and based on the BeatGAN model of the unsupervised learning method, we propose a new model by changing the kernel structure, adding the convolutional filter layer and the transposed convolutional filter layer to improve anomaly detection performance. In addition, by utilizing the generative features of the proposed model to implement and apply a retraining algorithm that generates new data and uses it for training, it was confirmed that the proposed model had the highest performance compared to the original BeatGAN models and other unsupervised learning model like Iforest and One Class SVM. Through this study, it was possible to suggest a method to improve the anomaly detection performance of proposed model while avoiding overfitting without additional cost in situations where training data are insufficient due to various factors such as sensor abnormalities and inspections in actual industrial sites.

• KSCI Review
• /
• v.28 no.1
• /
• pp.1-4
• /
• 2020

• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.5
• /
• pp.64-74
• /
• 2019

Although a non-rigid registration has high demands in clinical practice, it has a high computational complexity and it is very difficult for ensuring the accuracy and robustness of registration. This study proposes a method of applying a non-rigid registration to 3D magnetic resonance images of brain in an unsupervised learning environment by using a deep-learning network. A feature vector between two images is produced through the network by receiving both images from two different patients as inputs and it transforms the target image to match the source image by creating a displacement vector field. The network is designed based on a U-Net shape so that feature vectors that consider all global and local differences between two images can be constructed when performing the registration. As a regularization term is added to a loss function, a transformation result similar to that of a real brain movement can be obtained after the application of trilinear interpolation. This method enables a non-rigid registration with a single-pass deformation by only receiving two arbitrary images as inputs through an unsupervised learning. Therefore, it can perform faster than other non-learning-based registration methods that require iterative optimization processes. Our experiment was performed with 3D magnetic resonance images of 50 human brains, and the measurement result of the dice similarity coefficient confirmed an approximately 16% similarity improvement by using our method after the registration. It also showed a similar performance compared with the non-learning-based method, with about 10,000 times speed increase. The proposed method can be used for non-rigid registration of various kinds of medical image data.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.1
• /
• pp.1-12
• /
• 2023

It is very time-intensive to obtain data with labels on anomaly detection tasks for multivariate time series. Therefore, several studies have been conducted on unsupervised learning that does not require any labels. However, a well-done integrative survey has not been conducted on in-depth discussion of learning architecture and property for multivariate time series anomaly detection. This study aims to explore the characteristic of well-known architectures in anomaly detection of multivariate time series. Additionally, architecture was categorized by using top-down and bottom-up approaches. In order toconsider real-world anomaly detection situation, we trained models with dataset such as power grids or Cyber Physical Systems that contains realistic anomalies. From experimental results, we compared and analyzed the comprehensive performance of each architecture. Quantitative performance were measured using precision, recall, and F1 scores.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.291-295
• /
• 2018

한국어는 교착어적 특성이 강한 언어로, 교착어적 특성이 없는 영어 등의 언어와 달리 형태소의 수에 따라 조합 가능한 어절의 수가 매우 많으므로 어절 단위의 처리가 매우 어렵다. 따라서 어절을 더 작은 단위로 분해하는 전처리 단계가 요구되는데, 형태소 분석이 이를 위해 주로 사용되었다. 하지만 지도학습 방법을 이용한 형태소 분석 시스템은 다량의 학습 데이터가 요구되고, 비지도학습 방법을 이용한 형태소 분석은 성능에 큰 하락을 보인다. Byte Pair Encoding은 데이터를 압축하는 알고리즘으로, 이를 자연어처리 분야에 응용하면 비지도학습 방법으로 어절을 더 작은 단위로 분해할 수 있다. 본 연구에서는 한국어에 Byte Pair Encoding을 적용하는 두 가지 방법인 음절 단위 처리와 자모 단위 처리의 성능 및 특성을 정량적, 정성적으로 분석하는 방법을 제안하였다. 또한, 이 방법을 세종 말뭉치에 적용하여 각각의 알고리즘을 이용한 어절 분해를 실험하고, 그 결과를 어절 분해 정확도, 편향, 편차를 바탕으로 비교, 분석하였다.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.483-486
• /
• 2022

최근 이커머스 시장의 지속적인 성장으로 빠른 배송과 대용량 물류 처리를 위한 효율적 배송 시스템 마련의 필요성이 증가하고 있다. 본 연구에서는 도심 물류 거점에서의 현재 배송 물량 할당의 불균등 문제를 실무적 관점에서 정의하고, 비지도 학습 기반 클러스터링 기법을 통해 불균등 배송 할당 문제를 개선해 보고자 했다. 분석 결과 K-means++ 알고리즘 기반 클러스터링에서 최적화된 물량 할당에 대한 개선 가능성을 검증할 수 있었다. 향후 지형 정보, 교통량 등의 상세 변수를 추가하여 머신러닝 기반의 물류 배송 최적화를 위한 연구 영역을 확장할 수 있을 것으로 기대된다.