• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.213-216
• /
• 2021

기술이 발전함에 따라 악성코드 또한 함께 발전하여 보안 위협이 증가되고 있다. 특히 PowerShell 과 같은 스크립트 언어를 사용하여 포렌식이 어려운 Fileless 악성코드가 지속적으로 증가하고 있다. 이에 본 논문에서 PowerShell 데이터셋을 활용하여 기존 패턴은 탐지할 수 없는 한계점을 가진 시그니처 또는 휴리스틱 기반 탐지 기법을 보완하여 기존의 악성코드들을 학습 및 새로운 악성코드를 추측하는 것이 가능한 심층 학습 기술, Graph Convolutional Networks 과 자카드 유사도를 활용하여 기존의 방식에 비해 더 효율적으로 탐지를 이루어 내는지를 판단해보려한다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.113-118
• /
• 2023

딥러닝을 이용한 언어 모델은 다양한 분야에서 사용되고 있는데, 그 중 교육 분야에선 꾸준히 시험 문항을 자동으로 생성하려는 요구가 존재해 왔다. 그러나 빈칸 추론 문항, 그 중에서도 어구 단위 빈칸 추론 문항은 학습 및 평가 목적으로 널리 쓰이고 있지만, 이를 자동 생성하려는 연구는 상대적으로 드물다. 이에 본 연구에선 masked language modeling (MLM)을 이용한 난이도 조절이 가능한 어구 단위 빈칸 추론 문항 생성 시스템을 제안한다. 본 시스템은 정답 생성 모델의 attention 정보에 따라 지문 내 중요한 어구를 삭제해 오답을 생성하고, 동시에 어구의 삭제 비율을 조절함으로써 더 쉽거나 더 어려운 오답을 만들어낼 수 있다. 평가 결과, 제안한 시스템은 기존 접근법보다 정답과의 유사도가 최고 28.3% 낮았고, 또한 난이도 설정에 따라 쉬운 오답이 어려운 오답에 비해 유사도가 15.1% 낮아, 더 정답과 먼 뜻의 오답을 생성해내었다.

• Journal of KIISE:Databases
• /
• v.34 no.5
• /
• pp.396-408
• /
• 2007

After the Human Genome Project finished the sequencing of a human DNA sequence, the concerns on protein functions are increasing. Since the structures of proteins are conserved in divergent evolution, their functions are determined by their structures rather than by their amino acid sequences. Therefore, if similarities between two protein structures are observed, we could expect them to have common biological functions. So far, a lot of researches on protein structure alignment have been performed. However, most of them use RMSD(Root Mean Square Deviation) as a similarity measure with which it is hard to judge the similarity level of two protein structures intuitively. In addition, they retrieve only one result having the highest alignment score with which it is hard to satisfy various users of different purpose. To overcome these limitations, we propose a novel protein structure alignment algorithm based on MRPD(Maximum of Residue Pair Distance) and SG (Similarity Graph). MRPD is more intuitive similarity measure by which fast tittering of unpromising pairs of protein pairs is possible, and SG is a compact representation method for multiple alignment results with which users can choose the most plausible one among various users' needs by providing multiple alignment results without compromising the time to align protein structures.

• Journal of the Korea Concrete Institute
• /
• v.15 no.5
• /
• pp.705-714
• /
• 2003

Even though Korean peninsula is located in regions of moderate seismic risks, current seismic design provisions of the roadway bridge design code have adopted the AASHTO code which is based on the requirements for high seismic regions. The objective of this research is to investigate the seismic performance of circular reinforced concrete (RC) bridge piers with limited ductility, which may be desirable in low or moderate seismic regions, such as in Korea. Four test specimens were designed and constructed. The reference specimen was designed with longitudinal steel ratio as 1.01% and the confinement reinforcement ratio as 0.13% without considering earthquake, and three other test specimens were designed in accordance with a limited-ductility concept as 0.3% for the confinement steel ratio. This confinement ratio is 0.32 times of minimum lateral reinforcement specified in current seismic design provisions, and 2.3 times of lateral reinforcement required in nonseismic design provisions. The pseudo-dynamic test was carried out to evaluate the seismic performance of full-scale specimens in size of 1.2m diameter and 4.8m height. Judging from the experiment, the reference specimen was not satisfactory for the demand displacement ductility ${\mu}$=5.0, but three limited-ductility specimens appeared to have the displacement ductility of more than 5.

• Journal of Korea Water Resources Association
• /
• v.45 no.5
• /
• pp.445-454
• /
• 2012

The present study aims to investigate the long-term channel morphological changes derived from channelization, embankment and levee construction works in unregulated fluvial channel of the Jiseock River. Analyses of aerial photographs taken past (Year 1966) and recent (Year 2002) showed the temporally remarkable changes in channel planform such as channel shape, bar migration, vegetation encroachment in bar. During the period, the natural single threading changed into braided types together with decreasing sinuosity by 9.2%, increasing vegetation occupied bar ranged 97% of total bars area. Because such channel morphological changes are closely similar to those in dam downstream channels, we assume that both/either flow regime alteration and/or sediment transport discontinuity may be critical for the fixed channel and spread of vegetated bars even in unregulated river without dam reservoir upstream. We found more reduced frequency and magnitude of flooding water level comparing with past, but no significant alteration of inter annual water level variation. Bed material has been coarsened by 4~5 times and the riverbed has been degraded in overall channel but aggraded locally in conjunction reach of tributaries. The results indicates that reduced sediment dynamics in fluvial channel which derived by bed material coarsening, river bed degradation and unbalanced sediment transport capacity between tributary and mainstem can be a causal factor to trigger channel morphological changes even in unregulated rivers.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.3
• /
• pp.321-335
• /
• 2023

Monitoring sediment loads in natural rivers is the key process in river engineering, but it is costly and dangerous. In practice, suspended loads are directly measured, and total loads, which is a summation of suspended loads and bed loads, are estimated. This study proposes a real-time sediment discharge monitoring system using the horizontal acoustic Doppler current profiler (H-ADCP) and support vector regression (SVR). The proposed system is comprised of the SVR model for suspended sediment concentration (SVR-SSC) and for total loads (SVR-Q_TL), respectively. SVR-SSC estimates SSC and SVR-Q_TL mimics the modified Einstein procedure. The grid search with K-fold cross validation (Grid-CV) and the recursive feature elimination (RFE) were employed to determine SVR's hyperparameters and input variables. The two SVR models showed reasonable cross-validation scores (R²) with 0.885 (SVR-SSC) and 0.860 (SVR-Q_TL). During the time-series sediment load monitoring period, we successfully detected various sediment transport phenomena in natural streams, such as hysteresis loops and sensitive sediment fluctuations. The newly proposed sediment monitoring system depends only on the gauged features by H-ADCP without additional assumptions in hydraulic variables (e.g., friction slope and suspended sediment size distribution). This method can be applied to any ADCP-installed discharge monitoring station economically and is expected to enhance temporal resolution in sediment monitoring.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.7 no.3
• /
• pp.79-88
• /
• 2003

For the construction of PC bridge piers the implementation of 1992 seismic provisions, longitudinal steels were practically lap-spliced in the plastic hinge region. Experimental investigation was conducted ductility of evaluate the seismic earthquake-experienced reinforced concrete columns with 2,5 aspect ratio. Six test specimens were mode with test parameters: confinement ratios, lap splices, and retrofitting FRP materials. They were damaged under series of artificial earthquakes that could be compatible in Korean peninsula. Directly after the pseudo-dynamic test, damaged columns were retested under inelastic reversal cyclic loading simultaneously under a constant axial load, P＝0.1f$\_$ck/A$\_$g/. Residual seismic performance of damaged columns was evaluated and compared to that of the corresponding original columns. Test results show that PC bridge piers with lap-spliced longitudinal steels appeared to fail at low ductility. This was due to the debonding of the lap splice, which resulted from insufficient development of the longitudinal steels. The specimens externally wrapped with composite FRP straps in the potential plastic hinge region indicated significant improvement both in flexural strength and displacement ductility.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.8
• /
• pp.1049-1054
• /
• 2018

Keyword searching used in the past as a method of finding similar patents, and automated classification by machine learning is using in recently. Keyword searching is a method of analyzing data that is formalized through data refinement. While the accuracy for short text is high, long one consisted of several words like as document that is not able to analyze the meaning contained in sentences. In semantic analysis level, the method of automatic classification is used to classify sentences composed of several words by unstructured data analysis. There was an attempt to find similar documents by combining the two methods. However, it have a problem in the algorithm w the methods of analysis are different ways to use simultaneous unstructured data and regular data. In this paper, we study the method of extracting keywords implied in the document and using the LDA(Latent Semantic Analysis) method to classify documents efficiently without human intervention and finding similar patents.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.5
• /
• pp.173-179
• /
• 2008

This paper proposes an efficient 1:N matching scheme for fingerprint identification. Usually, in the minutiae-based matching scheme, fingerprint matching score could be calculated by analyzing geometrical similarity between minutiae from two fingerprints. To calculate the geometrical similarity between them, it is necessary to fingerprint align a fingerprint data with the other one. The final matching score is obtained by bidirectional matching in the common fingerprint matching scheme, because the similarity between two fingerprints varies with the result of alignments. The reliability of matching score by the bidirectional matching is better than by the unidirectional matching, but it takes two times comparing with unidirectional matching. To solve the problem, this paper proposes an efficient 1:N fingerprint matching scheme based on the distribution of bidirectional matching scores for the large fingerprints database. The experimental result shows the usefulness of the proposed scheme.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.2
• /
• pp.1054-1063
• /
• 2015

This study is to verify the effect of similarity of expanded-product, brand concepts of parent-brands and implicit theory of customers on the brand-extension evaluation, in order to reduce the risk of brand extension. First, this research documents how the similarity between the parent-brands and expanded-products affect brand-extension evaluations as a main effect variable. Second, this study examines the moderating effect of the brand concepts of parent-brands on the brand-extension evaluation. Third, this research documents how implicit theories regarding personality affect consumer evaluations about the brand-extension. The study assumes and tests that consumers in the group of incremental theorists are more accepting of brand-extensions than consumers in the group of entity theorists. The result figures out the implicit theory customers has some moderating effect on the evaluations, yet the direction of the effects is contrary to expectations.