• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.3
• /
• pp.523-533
• /
• 2014

It has been spent too much time to figure out the incident route when we are facing computer security incident. The incident often recurs moreover the damage is expanded because critical clues are lost while we are wasting time with hesitation. This paper suggests to build a Digital Evidence Map (DEM) in order to find out the incident cause speedy and accurately. The DEM is consist of the log chain which is a mesh relationship between machine data. And the DEM should be managed constantly because the log chain is vulnerable to various external facts. It could help handle the incident quickly and cost-effectively by acquainting it before incident. Thus we can prevent recurrence of incident by removing the root cause of it. Since the DEM has adopted artifacts in data as well as log, we could make effective response to APT attack and Anti-Forensic.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.369-384
• /
• 2021

In a personalized product recommendation system, when the amount of log data is large or sparse, the accuracy of model recommendation will be greatly affected. To solve this problem, a personalized product recommendation method using deep factorization machine (DeepFM) to analyze user behavior is proposed. Firstly, the K-means clustering algorithm is used to cluster the original log data from the perspective of similarity to reduce the data dimension. Then, through the DeepFM parameter sharing strategy, the relationship between low- and high-order feature combinations is learned from log data, and the click rate prediction model is constructed. Finally, based on the predicted click-through rate, products are recommended to users in sequence and fed back. The area under the curve (AUC) and Logloss of the proposed method are 0.8834 and 0.0253, respectively, on the Criteo dataset, and 0.7836 and 0.0348 on the KDD2012 Cup dataset, respectively. Compared with other newer recommendation methods, the proposed method can achieve better recommendation effect.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.4 no.1
• /
• pp.119-124
• /
• 2004

Real data as web log file tend to be incomplete. But we have to find useful knowledge from these for optimal decision. In web log data, many useful things which are hyperlink information and web usages of connected users may be found. The size of web data is too huge to use for effective knowledge discovery. To make matters worse, they are very sparse. We overcome this sparse problem using Markov Chain Monte Carlo method as multiple imputations. This missing value imputation changes spare web data to complete. Our study may be a useful tool for discovering knowledge from data set with sparseness. The more sparseness of data in increased, the better performance of MCMC imputation is good. We verified our work by experiments using UCI machine learning repository data.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.15-18
• /
• 2000

In this study, we improve the performance of a speech recognition system of visual information depending on lip movements. This paper focuses on the robustness of the word recognition system with the rotation, transition and scaling of the lip images. The different methods of lipreading have been used to estimate the stability of recognition performance. Especially, we work out the special system of the log-polar mapping, which is called Mellin transform with quasi RTS-invariant and related approaches to machine vision. The results of word recognition are reported with HMM (Hidden Markov Model) recognition system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1722-1737
• /
• 2019

We developed an insider threat detection model to be used by organizations that repeat tasks at regular intervals. The model identifies the best combination of different feature selection algorithms, unsupervised learning algorithms, and standard scores. We derive a model specifically optimized for the organization by evaluating each combination in terms of accuracy, AUC (Area Under the Curve), and TPR (True Positive Rate). In order to validate this model, a four-year log was applied to the system handling sensitive information from public institutions. In the research target system, the user log was analyzed monthly based on the fact that the business process is processed at a cycle of one year, and the roles are determined for each person in charge. In order to classify the behavior of a user as abnormal, the standard scores of each organization were calculated and classified as abnormal when they exceeded certain thresholds. Using this method, we proposed an optimized model for the organization and verified it.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.8
• /
• pp.99-106
• /
• 2012

In recent years, many researchers studies to promote the welfare of disabled people using IT technology. In particular, their suggestions are used a lot of mobile sensor installed on the street. These systems are acquired and to store the data sent to the server over the network, and by analyzing the users life log to judge of their risk state. In particular, persons with disabilities are exposed to various risks. So, they must need to the guardians if he go out. Thus, this study is a method for alleviating these so much pressure to smart appliances and impaired life log analysis system.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2004.04a
• /
• pp.160-165
• /
• 2004

At this research, we examined probability of light bulb's life span value and prediction on purpose to inquire out the span of repeat velocity as fracture probability by executing the fatigue test, which is considered property of Tungsten filament's thermal fatigue used as an automobile bulb. As a result we can confirm what the most suitable solution is weibull distribution and log normal distribution. Tungsten filament's span gets longer as the fatigue repeat velocity gets shorter And, repeat span is about 15%~40% shorter than sequence life span.

• MALSORI
• /
• no.49
• /
• pp.95-105
• /
• 2004

An automatic pronunciation correction system provides learners with correction guidelines for each mispronunciation. For this purpose we develope a speech recognizer which automatically classifies pronunciation errors when Koreans speak a foreign language. In order to develope the methods for automatic assessment of pronunciation quality, we propose a language model based score as a machine score in the speech recognizer. Experimental results show that the language model based score had higher correlation with human scores than that obtained using the conventional log-likelihood based score.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.1169-1171
• /
• 2007

인터넷 사용의 증가 및 정보보호에 대한 의식의 증가로 인하여 누가, 언제, 어떻게 해당 사이트를 이용 하였는가 뿐만 아니라 어떤 침해 사고를 일으키고 있느냐에 대한 이슈도 증가하고 있다. 따라서 본 논문에서는 방화벽 원시로그를 기계학습기법을 이용하여 보다 빠르게 방화벽 원시로그의 침해사고에 대한 지능형 모델을 제안한다.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2001.10a
• /
• pp.209-213
• /
• 2001

The reliability data analysis for components of CNC machining center is studied in this paper. The failure data of mechanical part is analyzed by Exponential, Weibull, and Log-normal distributions. And then, the optimum failure distribution model is selected by goodness of fit test. The reliability data analysis program is developed with ASP language to use on the Internet. The failure rate, MTBF, life, and failure mode of mechanical parts are estimated and searched by this program. The failure data and analysis results are stored in the database.