• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.229-247
• /
• 2023

With the increasing emphasis on anomaly detection across various fields, diverse anomaly detection algorithms have been developed for various data types and anomaly patterns. However, the performance of anomaly detection algorithms is generally evaluated on publicly available datasets, and the specific performance of each algorithm on anomalies of particular types remains unexplored. Consequently, selecting an appropriate anomaly detection algorithm for specific analytical contexts poses challenges. Therefore, in this paper, we aim to investigate the types of anomalies and various attributes of data. Subsequently, we intend to propose approaches that can assist in the selection of appropriate anomaly detection algorithms based on this understanding. Specifically, this study compares the performance of anomaly detection algorithms for four types of anomalies: local, global, contextual, and clustered anomalies. Through further analysis, the impact of label availability, data quantity, and dimensionality on algorithm performance is examined. Experimental results demonstrate that the most effective algorithm varies depending on the type of anomaly, and certain algorithms exhibit stable performance even in the absence of anomaly-specific information. Furthermore, in some types of anomalies, the performance of unsupervised anomaly detection algorithms was observed to be lower than that of supervised and semi-supervised learning algorithms. Lastly, we found that the performance of most algorithms is more strongly influenced by the type of anomalies when the data quantity is relatively scarce or abundant. Additionally, in cases of higher dimensionality, it was noted that excellent performance was exhibited in detecting local and global anomalies, while lower performance was observed for clustered anomaly types.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.302-313
• /
• 2023

This paper aims to develop a framework that can fully automate the quality management of training data used in large-scale Artificial Intelligence (AI) models built by the Ministry of Science and ICT (MSIT) in the 'AI Hub Data Dam' project, which has invested more than 1 trillion won since 2017. Autonomous driving technology using AI has achieved excellent performance through many studies, but it requires a large amount of high-quality data to train the model. Moreover, it is still difficult for humans to directly inspect the processed data and prove it is valid, and a model trained with erroneous data can cause fatal problems in real life. This paper presents a dataset reconstruction framework that removes abnormal data from the constructed dataset and introduces strategies to improve the performance of AI models by reconstructing them into a reliable dataset to increase the efficiency of model training. The framework's validity was verified through an experiment on the autonomous driving dataset published through the AI Hub of the National Information Society Agency (NIA). As a result, it was confirmed that it could be rebuilt as a reliable dataset from which abnormal data has been removed.

• Journal of Internet Computing and Services
• /
• v.20 no.6
• /
• pp.137-142
• /
• 2019

This paper Various services exist to detect and monitor abnormal event. However, most services focus on fires and gas leaks. so It is impossible to prevent and respond to emergency situations for the elderly and severely disabled people living alone. In this study, AI model is designed and compared to detect abnormal event of heart rate signal which is considered to be the most important among various bio signals. Specifically, electrocardiogram (ECG) data is collected using Physionet's MIT-BIH Arrhythmia Database, an open medical data. The collected data is transformed in different ways. We then compare the trained AI model with the modified and ECG data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.11
• /
• pp.1608-1614
• /
• 2022

As competition between universities intensifies due to the recent decrease in the number of students, it is recognized as an essential task of universities to predict students who are underperforming at an early stage and to make various efforts to prevent dropouts. For this, a high-performance model that accurately predicts student performance is essential. This paper proposes a method to improve prediction performance by removing or amplifying abnormal data in a classification prediction model for identifying underperforming students. Existing anomaly data processing methods have mainly focused on deleting or ignoring data, but this paper presents a criterion to distinguish noise from change indicators, and contributes to improving the performance of predictive models by deleting or amplifying data. In an experiment using open learning performance data for verification of the proposed method, we found a number of cases in which the proposed method can improve classification performance compared to the existing method.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.837-840
• /
• 2021

안정적인 서버 운영을 위해 이상 패턴 및 개체를 식별하는 이상탐지 연구가 활발하게 연구되어 오고 있다. 이상탐지의 대표적인 예로 서버의 사용량 증가를 꼽을 수 있지만, 실제 이상 데이터 수집 및 현상의 재현이 어렵다는 점은 해당 연구의 어려움으로 존재한다. 본 연구는 다양한 시나리오 기반의 부하테스트를 설계하고, 클라우드 환경에서 이상 데이터를 생성 및 수집하였다. 해당 데이터는 이상탐지에 대표적으로 사용되는 알고리즘의 성능을 비교 분석에 활용하였으며, 실험을 통해 각 알고리즘의 신뢰 수준을 확인하였다. 이는 다양한 서버 운영 환경에 적합한 알고리즘을 채택하는데 활용 가능하며, 결과적으로 안정적이고 효율적인 서버 운영에 기여할 수 있을 것으로 사료된다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.5
• /
• pp.47-53
• /
• 2023

With the advent of the 4th Industrial Revolution, diverse and a large amount of data has been accumulated now. The agricultural community has also collected environmental data that affects the growth of crops in smart farms or open fields with sensors. Environmental data has different features depending on where and when they are measured. Studies have been conducted using collected agricultural data to predict growth and yield with statistics and artificial intelligence. The results of these studies vary greatly depending on the data on which they are based. So, studies to enhance data quality have also been continuously conducted for performance improvement. A lot of data is required for high performance, but if there are outlier or missing values in the data, it can greatly affect the results even if the amount is sufficient. So, adjustment of outlier and missing values is essential in the data preprocessing. Therefore, this paper integrates data collected from actual farms and proposes a adjustment system for outlier and missing values based on it.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.856-858
• /
• 2024

VR 기술의 발전은 다양한 분야에서 사용자에게 몰입감 있는 가상 현실 경험을 제공하지만, VR기기 내부에 사용자의 생체 데이터 및 금융정보와 같은 민감한 정보들이 저장되어 새로운 보안 문제를 야기하고 있다. 이에 따라 PIN, 패스워드 등과 같은 기존의 인증 방식이 VR 기기에 적용되고 있지만 이들은 shoulder-surfing attack 공격 취약하며 VR 환경에서 사용하기에 불편한 인터페이스를 가지고 있다. 따라서 본 논문에서는 이상 탐지 모델을 활용하여 외부 추론 공격에 강인하며 VR 환경에 적합한 사용자 행위 기반의 VR기기 사용자 인증 모델을 구현한다. 특정 task를 수행하는 동안 사용자의 행위 데이터를 수집 및 feature 데이터를 추출하고, 정상으로 라벨링 된 사용자의 데이터로 이상 탐지 머신러닝 모델들을 학습 후 정상 데이터와 비정상 데이터를 이용하여 인증 모델의 성능을 평가하였다. OC-SVM이 87.72%의 F1-score로 세 모델 중 가장 높은 성능을 보임을 확인하였으며, 향후 인증 모델 성능 향상을 위한 계획을 제시하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.55-57
• /
• 2005

과학 분야의 발전에 따라 처리해야 하는 데이터의 양도 급격하게 증가하여 기가바이트, 테라바이트 혹은 페타바이트 이상이 되었다. 이렇게 큰 단위의 데이터를 로컬영역에서 처리하기에는 무리가 있다. 본 논문에서는 테라바이트 혹은 페타 바이트 이상의 데이터를 처리하고 관리하는 방안과 데이터의 사용방법에 대해서 논의 하겠다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.161-162
• /
• 2019

본 논문에서는 변압기의 이상 신호 데이터를 원거리에서 축적하기 위한 이상 신호 진단 시스템을 제안한다. 이 시스템은 원격으로 변압기의 이상 신호를 수집하여 변압기의 이상을 알아낼 수 있다. 이 시스템은 상용된 통신 모듈의 느린 통신 속도를 고려하여 이상 신호를 RMS-DC 값으로 변환하는 작업을 추가하였다.

• Korean Journal of Remote Sensing
• /
• v.28 no.5
• /
• pp.573-586
• /
• 2012

Point clouds acquired by a LIDAR(Light Detection And Ranging, also LADAR) system often contain erroneous points called outliers seeming not to be on physical surfaces, which should be carefully detected and eliminated before further processing for applications. Particularly in case of LIDAR systems employing with a Gieger-mode array detector (GmFPA) of high sensitivity, the outlier ratio is significantly high, which makes existing algorithms often fail to detect the outliers from such a data set. In this paper, we propose a method to discriminate outliers from a point cloud with high outlier ratio acquired by a GmFPA LIDAR system. The underlying assumption of this method is that a meaningful targe surface occupy at least two adjacent pixels and the ranges from these pixels are similar. We applied the proposed method to simulated LIDAR data of different point density and outlier ratio and analyzed the performance according to different thresholds and data properties. Consequently, we found that the outlier detection probabilities are about 99% in most cases. We also confirmed that the proposed method is robust to data properties and less sensitive to the thresholds. The method will be effectively utilized for on-line realtime processing and post-processing of GmFPA LIDAR data.