• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1301-1316
• /
• 2024

In modern warfare, missiles play a pivotal role but typically spend the majority of their lifecycle in long-term storage or standby mode, making it difficult to detect failures. Preemptive detection of missiles that will fail is crucial to preventing severe consequences, including safety hazards and mission failures. This study proposes a contamination-based stacking ensemble model, employing the local outlier factor (LOF), to detect such missiles. The proposed model creates multiple base LOF models with different contamination values and combines their anomaly scores to achieve a robust anomaly detection. A comparative performance analysis was conducted between the proposed model and the traditional single LOF model, using production-related inspection data from missiles deployed in the military. The experimental results showed that, with the contamination parameter set to 0.1, the proposed model exhibited an increase of approximately 22 percentage points in accuracy and 71 percentage points in F1-score compared to the single LOF model. This approach enables the preemptive identification of potential failures, undetectable through traditional statistical quality control methods. Consequently, it contributes to lower missile failure rates in real battlefield scenarios, leading to significant time and cost savings in the military industry.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.109-118
• /
• 2022

Workers' anomalous trajectories allow us to detect emergency situations in the workplace, such as accidents of workers, security threats, and fire. In this work, we develop a scheme to detect abnormal trajectories of workers using the edit distance on real sequence (EDR) and density method. Our anomaly detection scheme consists of two phases: offline phase and online phase. In the offline phase, we design a method to determine the algorithm parameters: distance threshold and density threshold using accumulated trajectories. In the online phase, an input trajectory is detected as normal or abnormal. To achieve this objective, neighbor density of the input trajectory is calculated using the distance threshold. Then, the input trajectory is marked as an anomaly if its density is less than the density threshold. We also evaluate performance of the proposed scheme based on the MIT Badge dataset in this work. The experimental results show that over 80 % of anomalous trajectories are detected with a precision of about 70 %, and F1-score achieves 74.68 %.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• Smart Media Journal
• /
• v.13 no.2
• /
• pp.85-94
• /
• 2024

Artificial Intelligence of Things (AIoT), which combines AI and the Internet of Things (IoT), has recently gained popularity. Deep neural networks (DNNs) have achieved great success in many applications. Deploying complex AI models on embedded boards, nevertheless, may be challenging due to computational limitations or intelligent model complexity. This paper focuses on an AIoT-based system for smart sewing automation using edge devices. Our technique included developing a detection model and a decision tree for a sufficient testing scenario. YOLOv5 set the stage for our defective sewing stitches detection model, to detect anomalies and classify the sewing patterns. According to the experimental testing, the proposed approach achieved a perfect score with accuracy and F1score of 1.0, False Positive Rate (FPR), False Negative Rate (FNR) of 0, and a speed of 0.07 seconds with file size 2.43MB.

• The Journal of Bigdata
• /
• v.6 no.2
• /
• pp.29-38
• /
• 2021

Deviation of route in aviation safety management is a dangerous factor that can lead to serious accidents. In this study, the anomaly score is calculated by classifying the tracks through clustering and calculating the distance from the cluster center. The study was conducted by extracting tracks within 100 km of the airport from the ADS-B track data received for one year. The wake was vectorized using linear interpolation. Latitude, longitude, and altitude 3D coordinates were used. Through PCA, the dimension was reduced to an axis representing more than 90% of the overall data distribution, and k-means clustering, hierarchical clustering, and PAM techniques were applied. The number of clusters was selected using the silhouette measure, and an abnormality score was calculated by calculating the distance from the cluster center. In this study, we compare the number of clusters for each cluster technique, and evaluate the clustering result through the silhouette measure.

• The Korean Nurse
• /
• v.30 no.3
• /
• pp.49-65
• /
• 1991

This study was carried out to assess the problems with the pregnant women, and check out the risk-factors in the high-risk pregnancies, using a simplified antepartum risk-scoring system, which was revised from Edwards' scoring system to be suitable for Korean situaition. This instrument was included 4 categories, demographic, obstetric, medical and miscellaneous factors. This survey was based on the 1300 pregnant women who were admitted, $x^2$-test, F-test, Pearsons correation, using statistical package SAS in NAS computer system, KIST. The results of the study were as follows; 1. 1313 infants were deliveried of these 560 infants(42.7%) were born to mothers with risk-scores > 7, and 753 infants(57.3%) were born to mothers risk-scores <7. 2. Maternal age" parity, education level, of the demographic factors were significant relation statistically to identify the high risk pregnancies($X^2$=20.88, 42.87, 15.60 P < 0.01). 3. C-section, post term, incompetent cervix, uterine anomaly, polyhydramnios, congenital anomaly, sensitized RH negative, abortion, preeclampsia, excessive size infant, premature, low birth weight infanl, abnormal presentation, perinatal loss, multiple pregnancy, of the obstetric factors were significant relation statistically to identify the high risk-pregnancies. ($X^2$ = 175.96, 87.5, 16.28, 21.78, 9.46, 8. 10, 6.75, 22.9, 64.84, 6.93, 361.43, 185.55, 78.65, 45.52, P < 0.01). 4. Abnormal nutrition, anemia, UTI, other medicalcondition(pulmonary disease, severe influenza), heart disease, V.D., of the miscellaneous and medical factors, were significant relation statistically to identify the high risk-pregnancies. 5. Premature, low birth weight infant, contracted pelvis, abnormal presentation, of the risk factors were significantly related with Apgar score at 1 '||'&'||' 5 minute after birth and neonatal body weight. 6. Apgar score at 1 '||'&'||' 5 minute after, birth and neonatal body weight were significantly negative correlated with risk-score. 7. There were statistically significant difference between risk-score and Apgar score at 1 '||'&'||' 5 minute after birth, 3 group(0-3, 4-6, above 7), and neonatal body weight, 2 group(below 2.5kg, the other group) (F=104.65, 96.61, 284.92, P<0.01). 8. Apgar score at 1 '||'&'||' 5 minute after birth(below 7), and neonatal body weight(below 2.5kg), were significant relation statistically with risk score.($x^2$=65.99, 60.88, 177.07, P<0.01) were 60.8 %, 60% . 9. Correct classifications of morbid infants(l '||'&'||' 5 minute Apgar score < 7) were 77.8%, 83.8% and that of nonmorbid infants(l '||'&'||' 5 minute Apgar score > 7) were 60.8%, 60%. 10. There were statistically significant difference between dislribution of maternal risk-score among the morbid infants(l '||'&'||' 5 minute Apgar score < 7) and non morbid infants(l '||'&'||' 5 minute Apgar score> 7) ($x^2$=64.8, 58.8, P < 0.001). 11. There were statistically significant difference between distribution of morbid infants(l '||'&'||' 5 minute Apgar score < 7) and fetal death. 12. The predictivity for classifying high.risk cases was 12 % and for classifying low-risk cases was 98.3 % in 5 minute Apgar score. Suggestions for further studies are as follows; 1. Contineous prospective studies, using this newly revised scoring system are strongly recommended in the stetric service. 2. Besides risk facto~s used in this study, assessmenl of risks by factors in another scoring system and paralled studies related to perinatal outcome are strongly recommended.

• Clinical and Experimental Pediatrics
• /
• v.60 no.11
• /
• pp.353-358
• /
• 2017

Purpose: To develop and evaluate a simple screening tool to assess hearing loss in newborns. A derived score was compared with the standard clinical practice tool. Methods: This cohort study was designed to screen the hearing of newborns using transiently evoked otoacoustic emission and auditory brain stem response, and to determine the risk factors associated with hearing loss of newborns in 3 tertiary hospitals in Northern Thailand. Data were prospectively collected from November 1, 2010 to May 31, 2012. To develop the risk score, clinical-risk indicators were measured by Poisson risk regression. The regression coefficients were transformed into item scores dividing each regression-coefficient with the smallest coefficient in the model, rounding the number to its nearest integer, and adding up to a total score. Results: Five clinical risk factors (Craniofacial anomaly, Ototoxicity, Birth weight, family history [Relative] of congenital sensorineural hearing loss, and Apgar score) were included in our COBRA score. The screening tool detected, by area under the receiver operating characteristic curve, more than 80% of existing hearing loss. The positive-likelihood ratio of hearing loss in patients with scores of 4, 6, and 8 were 25.21 (95% confidence interval [CI], 14.69-43.26), 58.52 (95% CI, 36.26-94.44), and 51.56 (95% CI, 33.74-78.82), respectively. This result was similar to the standard tool (The Joint Committee on Infant Hearing) of 26.72 (95% CI, 20.59-34.66). Conclusion: A simple screening tool of five predictors provides good prediction indices for newborn hearing loss, which may motivate parents to bring children for further appropriate testing and investigations.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.41-43
• /
• 2024

본 연구에서는 Vision Transformer 기반의 Anomaly Detection and Localization (VT-ADL) 모델에 초점을 맞추고, 손실 함수의 변경이 MVTec 데이터셋에 대한 이상 검출 및 지역화 성능에 미치는 영향을 비교 분석한다. 기존의 손실 함수를 KL Divergence와 Log-Likelihood Loss의 조합인 VAE Loss로 대체하여, 성능 변화를 심층적으로 조사했다. 실험을 통해 VAE Loss로의 전환은 VT-ADL 모델의 이상 검출 능력을 현저히 향상시키며, 특히 PRO-score에서 기존 대비 약 5%의 개선을 보였다는 점을 확인하였다. 이러한 결과는 손실 함수의 최적화가 VT-ADL 모델의 전반적인 성능에 중요한 영향을 미칠 수 있음을 시사한다. 또한, 이 연구는 Vision Transformer 기반 모델의 이상 검출과 지역화 작업에 있어서 손실 함수 선택의 중요성을 강조하며, 향후 관련 연구에 유용한 기준을 제공할 수 있을 것으로 기대된다.

• Journal of Navigation and Port Research
• /
• v.42 no.4
• /
• pp.277-282
• /
• 2018

Recently, the Vessel Traffic Service (VTS) coverage has expanded to include coastal areas following the increased attention on vessel traffic safety. However, it has increased the workload on the VTS operators. In some cases, when the traffic volume increases sharply during the rush hour, the VTS operator may not be aware of the risks. Therefore, in this paper, we proposed a new method to recognize ship movement anomalies automatically to support the VTS operator's decision-making. The proposed method generated traffic pattern model without any category information using the unsupervised learning algorithm.. The anomaly score can be calculated by classification and comparison of the trained model. Finally, we reviewed the experimental results using a ship-handling simulator and the actual trajectory data to verify the feasibility of the proposed method.

• Convergence Security Journal
• /
• v.18 no.5_1
• /
• pp.45-51
• /
• 2018

As data utilization and importance becomes important, data-related accidents and damages are gradually increasing. Especially, insider threats are the most harmful threats. And these insider threats are difficult to detect by traditional security systems, so rule-based abnormal behavior detection method has been widely used. However, it has a lack of adapting flexibly to changes in new attacks and new environments. Therefore, in this paper, we propose an adaptive anomaly movement detection framework based on a statistical Markov model to detect insider threats in advance. This is designed to minimize false positive rate and false negative rate by adopting environment factors that directly influence the behavior, and learning data based on statistical Markov model. In the experimentation, the framework shows good performance with a high F2-score of 0.92 and suspicious behavior detection, which seen as a normal behavior usually. It is also extendable to detect various types of suspicious activities by applying multiple modeling algorithms based on statistical learning and environment factors.