• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.2-4
• /
• 2023

Sequence data plays an important role in the field of intelligence, especially for industrial control, traffic control and other aspects. Finding abnormal parts in sequence data has long been an application field of AI technology. In this paper, we propose an anomaly detection method for sequence data using a diffusion model. The diffusion model has two major advantages: interpretability derived from rigorous mathematical derivation and unrestricted selection of backbone models. This method uses the diffusion model to predict and reconstruct the sequence data, and then detects the abnormal part by comparing with the real data. This paper successfully verifies the feasibility of the diffusion model in the field of anomaly detection. We use the combination of MLP and diffusion model to generate data and compare the generated data with real data to detect anomalous points.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1887-1898
• /
• 2018

In this paper, a method to classify insider threat activity is introduced. The internal threats help detecting anomalous activity in the procedure performed by the user in an organization. When an anomalous value deviating from the overall behavior is displayed, we consider it as an inside threat for classification as an inside intimidator. To solve the situation, Markov Chain Model is employed. The Markov Chain Model shows the next state value through an arbitrary variable affected by the previous event. Similarly, the current activity can also be predicted based on the previous activity for the insider threat activity. A method was studied where the change items for such state are defined by a transition probability, and classified as detection of anomaly of the inside threat through values for a probability variable. We use the properties of the Markov chains to list the behavior of the user over time and to classify which state they belong to. Sequential data sets were generated according to the influence of n occurrences of Markov attribute and classified by machine learning algorithm. In the experiment, only 15% of the Cert: insider threat dataset was applied, and the result was 97% accuracy except for NaiveBayes. As a result of our research, it was confirmed that the Markov Chain Model can classify insider threats and can be fully utilized for user behavior classification.

• Journal of Chest Surgery
• /
• v.20 no.4
• /
• pp.695-705
• /
• 1987

This report provides follow - up data on 37 patients, aged 7 days to 25 years [median: 6.5 months], who underwent repair of total anomalous pulmonary venous connection at Seoul national University Hospital between May, 1978 and June, 1987. The patients were 22 males and 17 females and the sex ratio was 1.6 to 1, showing a male predominance. Sixteen patients had supracardiac, 13 cardiac, 3 infracardiac and 5 had a mixed type. The duration of follow up was from 1 month to 60 months [median: 14 months] There were eight early and one late deaths, and the overall mortality was 24%. The deaths during 1 year of life were eight [89%] and only one death [11%] occurred above 1 year of age. The mortality of cardiac type was unusually high, accounting for 56 percent of the total death, which was probably due to the preoperative poor clinical condition such as pulmonary edema and congestive heart failure. The major cause of death was the perioperative myocardial failure, and the survival was closely related to the preoperative clinical status, age and moderately elevated pulmonary arterial pressure, the sign of the elevated pulmonary vascular resistance and pulmonary venous obstruction. Early diagnosis and early application of surgical intervention is essential to the improved postoperative survival

• Convergence Security Journal
• /
• v.12 no.4
• /
• pp.77-84
• /
• 2012

The control network has been operated in a closed. But it changes to open to external for business convenience and cooperation with several organizations. As the way of connecting with user extends, the risk of control network gets high. Thus, in this paper, proposed the technique of an anomalous event detection using white-list for control network security and minimizing the cyber threats. The proposed method can be collected and cataloged of only normal data from traffic of internal network, control network and field devices. Through way to check the this situation, we can separate normal and abnormal behavior.

• The Bulletin of The Korean Astronomical Society
• /
• v.42 no.2
• /
• pp.55.5-56
• /
• 2017

Galaxy dynamics probes weak gravity at accelerations below the de Sitter scale of acceleration adS = cH, where c is the velocity of light and H is the Hubble parameter. Low and high redshift galaxies hereby offer a novel probe of weak gravity in an evolving cosmology, satisfying H(z) = H0(1 + A(6z + 12z^2 +12z^3+ 6z^4+ (6/5)z^5)/(1 + z) with baryonic matter content A sans tension to H0 in surveys of the Local Universe. Galaxy rotation curves show anomalous galaxy dynamics in weak gravity aN < adS across a transition radius r beyond about 5 kpc for galaxy mass of 1e11 solar mass. where aN is the Newtonian acceleration based on baryonic matter content. We identify this behavior with a holographic origin of inertia from entanglement entropy, that introduces a C0 onset across aN=adS with asymptotic behavior described by a Milgrom parameter satisfying a0=omega/(2pi), where omega=sqrt(1-q)H is a fundamental eigenfrequency of the cosmological horizon. Extending an earlier confrontation with data covering 0.003 < aN/adS < 1 at redshift z about zero in Lellie et al. (2016), the modest anomalous behavior in the Genzel et al. sample at redshifts 0.854 < z <2.282 is found to be mostly due to clustering 0.36 < aN/adS < 1 close to the C0 onset to weak gravity and an increase of up to 65% in a0.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.1
• /
• pp.27-35
• /
• 2016

Black-box classifiers, such as artificial neural network and support vector machine, are a popular classifier because of its remarkable performance. They are applied in various fields such as inductive inferences, classifications, or regressions. However, by its characteristics, they cannot provide appropriate explanations how the classification results are derived. Therefore, there are plenty of actively discussed researches about interpreting trained black-box classifiers. In this paper, we propose a method to make a fuzzy logic-based classifier using extracted rules from the artificial neural network and support vector machine in order to interpret internal structures. As an object of classification, an anomalous propagation echo is selected which occurs frequently in radar data and becomes the problem in a precipitation estimation process. After applying a clustering method, learning dataset is generated from clusters. Using the learning dataset, artificial neural network and support vector machine are implemented. After that, decision trees for each classifier are generated. And they are used to implement simplified fuzzy logic-based classifiers by rule extraction and input selection. Finally, we can verify and compare performances. With actual occurrence cased of the anomalous propagation echo, we can determine the inner structures of the black-box classifiers.

• Journal of Pharmaceutical Investigation
• /
• v.30 no.3
• /
• pp.207-211
• /
• 2000

The objective of this work is to investigate the effect of molecular weight of poly(ethylene oxide) (PEO) and release medium on the release of nifedipine (NP) from PEO tablets containing NP and to get some mechanistic insights into the release of NP. The tablets containing NP were prepared by direct compression, using a flat-faced punch and die. The molecular weights of PEOs used were 200K, 900K, 2000K and 7,000K. The release kinetics were studied for 24 hours in aqueous ethanol solution, using a dissolution tester at $36.5^{\circ}C$ and 100 rpm. Drug release rate increased, as the concentration of ethanol in the dissolution medium increased, due to the increased solubility of NP. As the molecular weight of PEO increased, release rate decreased, due to the slower swelling and dissolution of PEO. The power values obtained by fitting data to the power law expression $(M_t/M_{\infty}=kt^n)$ indicated that, at low ethanol concentration, the release of NP is governed by anomalous diffusion. However, as the ethanol concentration increases, diffusional release becomes to prevail over anomalous or zero-order release. Overall, these results provided some insights into the release of NP from PEO tablet.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.127-136
• /
• 2024

This study explores the correlation between system anomalies and large-scale logs within the Spark cluster environment. While research on anomaly detection using logs is growing, there remains a limitation in adequately leveraging logs from various components of the cluster and considering the relationship between anomalies and the system. Therefore, this paper analyzes the distribution of normal and abnormal logs and explores the potential for anomaly detection based on the occurrence of log templates. By employing Hadoop and Spark, normal and abnormal log data are generated, and through t-SNE and K-means clustering, templates of abnormal logs in anomalous situations are identified to comprehend anomalies. Ultimately, unique log templates occurring only during abnormal situations are identified, thereby presenting the potential for anomaly detection.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.159-172
• /
• 2024

The number of healthcare products available for pets has increased in recent times, which has prompted active research into wearable devices for pets. However, the data collected through such devices are limited by outliers and missing values owing to the anomalous and irregular characteristics of pets. Hence, we propose pet behavior recognition based on a hybrid one-dimensional convolutional neural network (CNN) and long short- term memory (LSTM) model using pet wearable devices. An Arduino-based pet wearable device was first fabricated to collect data for behavior recognition, where gyroscope and accelerometer values were collected using the device. Then, data augmentation was performed after replacing any missing values and outliers via preprocessing. At this time, the behaviors were classified into five types. To prevent bias from specific actions in the data augmentation, the number of datasets was compared and balanced, and CNN-LSTM-based deep learning was performed. The five subdivided behaviors and overall performance were then evaluated, and the overall accuracy of behavior recognition was found to be about 88.76%.

• Publications of The Korean Astronomical Society
• /
• v.32 no.1
• /
• pp.97-99
• /
• 2017

Our understanding of dust emission, interaction, and evolution, is evolving. In recent years, electric dipole emission by spinning dust has been suggested to explain the anomalous microwave excess (AME), appearing between 10 and 90 Ghz. The observed frequencies suggest that spinning grains should be on the order of 10nm in size, hinting at polycyclic aromatic hydrocarbon molecules (PAHs). We present data from the AKARI/Infrared Camera (IRC) due to its high sensitivity to the PAH bands. By inspecting the IRC data for a few AME regions, we find a preliminary indication that regions well-fitted by a spinning-dust model have a higher $9{\mu}m$ than $18{\mu}m$ intensity vs. non-spinning-dust regions. Ongoing efforts to improve the analysis by using DustEM and including data from the AKARI Far Infrared Surveyor (FIS), IRAS, and Planck High Frequency Instrument (HFI) are described.