• Smart Media Journal
• /
• v.13 no.4
• /
• pp.23-32
• /
• 2024

Large-scale fires and their consequential damages are becoming increasingly common, but confidence in fire detection systems is waning. Recently, widely-used chemical fire detectors frequently generate lots of false alarms, while video-based deep learning fire detection is hampered by its time-consuming and expensive nature. To tackle these issues, this study proposes a fire detection model utilizing an autoencoder approach. The objective is to minimize false alarms while achieving swift and precise fire detection. The proposed model, employing an autoencoder methodology, can exclusively learn from normal data without the need for fire-related data, thus enhancing its adaptability to diverse environments. By amalgamating data from five distinct sensors, it facilitates rapid and accurate fire detection. Through experiments with various hyperparameter combinations, the proposed model demonstrated that out of 14 scenarios, only one encountered false alarm issues. Experimental results underscore its potential to curtail fire-related losses and bolster the reliability of fire detection systems.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.1109-1112
• /
• 2006

DDoS(Distributed Denial-of-Service) 공격은 인터넷 침해가운데 가장 위협적인 공격들 중 하나이며 이러한 공격을 실시간으로 탐지하기 위한 연구는 활발히 이루어져 왔다. 하지만 기존의 탐지 메커니즘이 가지고 있는 높은 오탐지율은 여전히 보완해야할 과제로 남아 있다. 따라서 본 논문에서는 DDoS공격 탐지의 근거로 사용된 기존의 트래픽 볼륨(traffic volume), 엔트로피(entropy), 그리고 카이제곱(chi-square)을 이용한 비정상 행위탐지(Anomaly detection)방식의 침임탐지시스템이 가지는 오탐지율(false alarm rate)을 개선할 수 있는 방안을 제안한다. 또한 공격 탐지 시 프로토콜, TCP 플래그(flag), 그리고 포트 번호를 이용하여 네트워크 관리자에게 보다 자세한 공격 정보를 제공함으로써 효율적으로 공격에 대처할 수 있는 시스템을 설계한다.

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 2006.06a
• /
• pp.219-222
• /
• 2006

In this paper an intrusion detection system technique of wireless Ad Hoc network is explained and the advantage of making them work in IEEE 802.15.4/ZigBee wireless standard is also discussed. The methodology that is mentioned here is intrusion detection architecture based on a local intrusion database [1]. An ad hoc network is a collection of nodes that is connected through a wireless medium forming rapidly changing topologies. Due to increased connectivity (especially on the Internet), and the vast spectrum of financial possibilities that are opening up, more and more systems are subject to attack by intruders. An ideal IDS should able to detect an anomaly caused by the intruders quickly so that the misbehaving node/nodes can be identified and appropriate actions (e.g. punish or avoid misbehaving nodes) can be taken so that further damage to the network is minimized

• 한국IT서비스학회:학술대회논문집
• /
• 2009.05a
• /
• pp.357-360
• /
• 2009

본 연구는 데이터마이닝 기법을 이용하여 건강보험청구료에 있어서 이상정도가 심한 요양기관을 탐지하고, 실제 의료영역에 적용하기 위한 시스템 개발을 목적으로 한다. 현재 건강보험 심사평가원의 이상탐지시스템은 평가대상이 되는 항목을 개별적으로 평가하고, 탐지된 기관의 선정 이유에 대한 근거제시가 부족한 단점을 가지고 있다. 따라서 본 연구에서는 항목을 종합적으로 평가할 수 있는 정량적 지표를 설계하고, 항목들의 상대적 중요도를 파악할 수 있도록 항목들에 대한 가중치 부여한다. 또한 지표에서 얻어진 값으로 등급을 구분하고, 의사결정나무기법(decision tree)를 이용하여 해석력을 높이는 방법을 제시한다.

• Journal of Positioning, Navigation, and Timing
• /
• v.11 no.2
• /
• pp.109-117
• /
• 2022

This paper proposes a method for detecting flight anomalies of drones through the difference between the command of flight controller (FC) and the navigation solution. If the drones make a flight normally, control errors generated by the difference between the desired control command of FC and the navigation solution should converge to zero. However, there is a risk of sudden change or divergence of control errors when the FC control feedback loop preset for the normal flight encounters interferences such as strong winds or navigation sensor abnormalities. In this paper, we propose the method with a deep neural network model that predicts the control error in the normal flight so that the abnormal flight state can be detected. The performance of proposed method was evaluated using the real-world flight data. The results showed that the method effectively detects anomalies in various situation.

• Journal of Internet Computing and Services
• /
• v.23 no.2
• /
• pp.97-105
• /
• 2022

Web services show a rapid evolution and integration to meet the increased users' requirements. Thus, web services undergo updates and may have performance degradation due to undetected faults in the updated versions. Due to these faults, many performances and regression anomalies in web services may occur in real-world scenarios. This paper proposed applying the deep learning model and innovative explainable framework to detect performance and regression anomalies in web services. This study indicated that upper bound and lower bound values in performance metrics provide us with the simple means to detect the performance and regression anomalies in updated versions of web services. The explainable deep learning method enabled us to decide the precise use of deep learning to detect performance and anomalies in web services. The evaluation results of the proposed approach showed us the detection of unusual behavior of web service. The proposed approach is efficient and straightforward in detecting regression anomalies in web services compared with the existing approaches.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.2
• /
• pp.313-316
• /
• 2022

In this paper, we implement a system that detects abnormalities in the charging data transmitted from the charger during the charging process of electric vehicles and controls them remotely. Using classification algorithms such as logistic regression, KNN, SVM, and decision trees, to do this, an analysis model is created that judges the data received from the charger as normal and abnormal. In addition, a model is created to determine the cause of the abnormality using the existing charging data based on the analysis of the type of charger abnormality. Finally, it is solved using unsupervised learning method to find new patterns of abnormal data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.605-607
• /
• 2023

Recent research has highlighted the effectiveness of Deep Learning (DL) techniques in automating the detection of lung sound anomalies. However, the available lung sound datasets often suffer from limitations in both size and balance, prompting DL methods to employ data preprocessing such as augmentation and transfer learning techniques. These strategies, while valuable, contribute to the increased complexity of DL models and necessitate substantial training memory. In this study, we proposed a streamlined and lightweight DL method but effectively detects lung sound anomalies from small and imbalanced dataset. The utilization of 1D dilated convolutional neural networks enhances sensitivity to lung sound anomalies by efficiently capturing deep temporal features and small variations. We conducted a comprehensive evaluation of the ICBHI dataset and achieved a notable improvement over state-of-the-art results, increasing the average score of sensitivity and specificity metrics by 2.7%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.4
• /
• pp.763-773
• /
• 2017

In recent years, personal information leakage and technology leakage accidents are frequently occurring. According to the survey, the most important part of this spill is the 'insider' within the organization, and the leakage of technology by insiders is considered to be an increasingly important issue because it causes huge damage to the organization. In this paper, we try to learn the normal behavior of employees using machine learning to prevent insider threats, and to investigate how to detect abnormal behavior. Experiments on the detection of abnormal behavior by implementing an Autoencoder composed of Recurrent Neural Network suitable for learning time series data among the neural network models were conducted and the validity of this method was verified.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.2
• /
• pp.145-152
• /
• 2020

In this paper, we propose the Crop Classification Management System (CCMS) that detects changes in growth environment to improve crop production rate. The CCMS consists of two modules. First, the Crop Classification Module (CCM) classifies crops through CNN. Second, the Farm Anomaly Detection Module (FADM) detects abnormal crops by comparing accumulated data of farms. The CCM recognizes crops currently grown on farms and sends them to the FADM, and the FADM picks up the weather data from the past to the present day of the farm growing the crops and applies them to the Nelson rules. The FADM uses the Nelson rules to find out weather data that has occurred and adjust farm conditions through IoT devices. The performance analysis of CCMS showed that the CCM had a crop classification accuracy of about 90%, and the FADM improved the estimated yield by up to about 30%. In other words, managing farms through the CCMS can help increase the yield of smart farms.