• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.414-417
• /
• 2011

단어 필터링 기법의 개발에 있어서 가장 큰 난제는 정상단어를 금지어로 인식하여 이를 차단하는 오탐지 문제이다. 이를 방지하기 위하여 다양한 문장에 대한 면밀하고 광범위한 검사가 필수적이나 일반적으로 쉽게 접근할 수 있는 데이터는 주로 단어 위주로 구성된 검증 데이터로 문장 또는 구절로 구성된 데이터의 양은 실제 검증에 활용하기에는 충분하지 못하다. 본 논문에서는 웹에서 수집한 방대한 양의 말뭉치 데이터에 반전역정열(Semi-Global Alignment)을 적용하여 주어진 금지어가 사용되거나 금지어와 유사한 단어가 존재하는 구간을 탐색함으로써 단어 필터링 시스템에서 범용적으로 사용될 수 있는 문장형 평가 데이터를 수집하는 시스템을 제안하고 해당 기법을 통하여 수집한 문장 단위 데이터를 분석하고 단어 필터링 시스템이 오탐지 방지를 위해 가져야할 요소들에 대하여 검토해 본다.

• The Journal of the Korea Contents Association
• /
• v.9 no.12
• /
• pp.113-120
• /
• 2009

Currently the verbal abuse in text message over on-line game is so serious. However we do not have any effective policy or technical tools yet. Till now in order to cope with this problem, the online game service providers have accumulated a set of forbidden words and applied this list on the textual word used in on-line game, which is called 'Swear filter'. But young on-line game players easily avoid this filtering method by coining another words which is not kept in the list. Especially Korean is very easy to make new variations of a vulgar word. In this paper, we propose one smart filtering algorithm to identify newly coined profanities. Important features of our method include the canonical form transformation of coined profanities, semi-global alignment between in the level of consonant and vowel units. For experiment, we have collected more than 1000 newly coined vulgar words in on-line gaming sites and tested these word against our methods. where our system have successfully filtered more than 90% of those newly coined vulgar words.

• Journal of KIISE:Computer Systems and Theory
• /
• v.37 no.6
• /
• pp.354-367
• /
• 2010

There are many methods for analyzing patterns in time series data. Although stock data represents a time series, there are few studies on stock pattern analysis and prediction. Since people believe that stock price changes randomly we cannot predict stock prices using a scientific method. In this paper, we measured the degree of the randomness of stock prices using Kolmogorov complexity, and we showed that there is a strong correlation between the degree and the accuracy of stock price prediction using our semi-global alignment method. We transformed the stock price data to quantized string sequences. Then we measured randomness of stock prices using Kolmogorov complexity of the string sequences. We use KOSPI 690 stock data during 28 years for our experiments and to evaluate our methodology. When a high Kolmogorov complexity, the stock price cannot be predicted, when a low complexity, the stock price can be predicted, but the prediction ratio of stock price changes of interest to investors, is 12% prediction ratio for short-term predictions and a 54% prediction ratio for long-term predictions.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.4
• /
• pp.23-30
• /
• 2004

Ontologies play an important role in the Semantic Web by providing well-defined meaning to ontology consumers. But as the ontologies are authored in a bottom-up distributed mimer, a large number of overlapping ontologies are created and used for the similar domains. Ontology sharing and reuse have become a distinguished topic, and ontology merging and alignment are the solutions for the problem. Ontology merging and alignment algorithms previously proposed detect conflicts between concepts by making use of only local syntactic information of concept names. And they depend only on a semi-automatic approach, which makes ontology engineers tedious. Consequently, the quality of merging and alignment tends to be unsatisfying. To remedy the defects of the previous algorithms, we propose a new algorithm for ontology merging and alignment which uses local and global semantic set of a concept. We evaluated our algorithm with several pairs of ontologies written in OWL, and achieved around 91% of precision in merging and alignment. We expect that, with the widespread use of web ontology, the need for ontology sharing and reuse ill become higher, and our proposed algorithm can significantly reduce the time required for ontology development. And also, our algorithm can easily be applied to various fields such as ontology mapping where semantic information exchange is a requirement.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.1
• /
• pp.105-125
• /
• 2024

Gliomas are the most common malignant brain tumor and cause the most deaths. Manual brain tumor segmentation is expensive, time-consuming, error-prone, and dependent on the radiologist's expertise and experience. Manual brain tumor segmentation outcomes by different radiologists for the same patient may differ. Thus, more robust, and dependable methods are needed. Medical imaging researchers produced numerous semi-automatic and fully automatic brain tumor segmentation algorithms using ML pipelines and accurate (handcrafted feature-based, etc.) or data-driven strategies. Current methods use CNN or handmade features such symmetry analysis, alignment-based features analysis, or textural qualities. CNN approaches provide unsupervised features, while manual features model domain knowledge. Cascaded algorithms may outperform feature-based or data-driven like CNN methods. A revolutionary cascaded strategy is presented that intelligently supplies CNN with past information from handmade feature-based ML algorithms. Each patient receives manual ground truth and four MRI modalities (T1, T1c, T2, and FLAIR). Handcrafted characteristics and deep learning are used to segment brain tumors in a Global Convolutional Neural Network (GCNN). The proposed GCNN architecture with two parallel CNNs, CSPathways CNN (CSPCNN) and MRI Pathways CNN (MRIPCNN), segmented BraTS brain tumors with high accuracy. The proposed model achieved a Dice score of 87% higher than the state of the art. This research could improve brain tumor segmentation, helping clinicians diagnose and treat patients.

• Steel and Composite Structures
• /
• v.22 no.2
• /
• pp.301-321
• /
• 2016

The objective of this paper is to investigate buckling behavior of composite laminated cylinders by using semi-analytical finite strip method. The shell is subjected to deformation-dependent loads which remain normal to the shell middle surface throughout the deformation process. The load stiffness matrix, which is responsible for variation of load direction, is also throughout the deformation process. The shell is divided into several closed strips with alignment of their nodal lines in the circumferential direction. The governing equations are derived based on the first-order shear deformation theory with Sanders-type of kinematic nonlinearity. Displacements and rotations of the shell middle surface are approximated by combining polynomial functions in the meridional direction and truncated Fourier series along with an appropriate number of harmonic terms in the circumferential direction. The load stiffness matrix, which is responsible for variation of load direction, is also derived for each strip and after assembling, global load stiffness matrix of the shell is formed. The numerical illustrations concern the pressure stiffness effect on buckling pressure under various conditions. The results indicate that considering pressure stiffness causes buckling pressure reduction which in turn depends on various parameters such as geometry and lay-ups of the shell.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.9
• /
• pp.911-915
• /
• 2008

Software birthmark is the inherent characteristics that can identify a program. In this paper, we propose a Java class theft detection technique based on static API traces of class files. We utilize control flow analysis to increase resilience, and we apply the semi-global alignment trace comparison algorithm to increase credibility. The credibility and resilience experiments for XML parsers show that our birthmark is more efficient than existing birthmarks.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.46 no.6
• /
• pp.479-486
• /
• 2018

In order to ensure reliable navigation performance of a lunar exploration rover, navigation algorithms using additional sensors such as inertial measurement units and cameras are essential on lunar surface in the absence of a global navigation satellite system. Unprecedentedly, Visual Odometry (VO) using a stereo camera has been successfully implemented at the US Mars rovers. In this paper, we estimate the 6-DOF pose of the lunar exploration rover from gray images of a lunar-like terrains. The proposed algorithm estimates relative pose of consecutive images by sparse image alignment based semi-direct VO. In order to overcome vulnerability to non-linearity of direct VO, we add adaptive motion prior weights calculated from a linear function of the previous pose to the optimization cost function. The proposed algorithm is verified in lunar-like terrain dataset recorded by Toronto University reflecting the characteristics of the actual lunar environment.

• Journal of KIISE:Software and Applications
• /
• v.36 no.9
• /
• pp.767-776
• /
• 2009

A software birthmark is a set of characteristics that are extracted from a program itself to detect code theft. A dynamic API birthmark is extracted from the run-time API call sequences of a program. The dynamic Windows API birthmarks of Tamada et al. are extracted from API call sequences during the startup period of a program. Therefore. the dynamic birthmarks cannot reflect characteristics of main functions of the program. In this paper. we propose a functional unit birthmark(FDAPI) that is defined as API call sequences recorded during the execution of essential functions of a program. To find out that some functional units of a program are copied from an original program. two FDAPIs are extracted by executing the programs with the same input. The FDAPIs are compared using the semi-global alignment algorithm to compute a similarity between two programs. Programs with the same functionality are compared to show credibility of our birthmark. Binary executables that are compiled differently from the same source code are compared to prove resilience of our birthmark. The experimental result shows that our birthmark can detect module theft of software. to which the existing birthmarks of Tamada et al. cannot be applied.

• Structural Engineering and Mechanics
• /
• v.58 no.1
• /
• pp.59-91
• /
• 2016

Laminated composite shells are commonly used in various engineering applications including aerospace and marine structures. In this paper, using semi-analytical finite strip method, the buckling behavior of laminated composite deep as well as thick shells of revolution under follower forces which remain normal to the shell is investigated. The stiffness caused by pressure is calculated for the follower forces subjected to external fibers in thick shells. The shell is divided into several closed strips with alignment of their nodal lines in the circumferential direction. The governing equations are derived based on first-order shear deformation theory which accounts for through thickness-shear flexibility. Displacements and rotations in the middle surface of shell are approximated by combining polynomial functions in the meridional direction as well as truncated Fourier series with an appropriate number of harmonic terms in the circumferential direction. The load stiffness matrix which accounts for variation of loads direction will be derived for each strip of the shell. Assembling of these matrices results in global load stiffness matrix which may be un-symmetric. Upon forming linear elastic stiffness matrix called constitutive stiffness matrix, geometric stiffness matrix and load stiffness matrix, the required elements for the second step analysis which is an eigenvalue problem are provided. In this study, different parameter effects are investigated including shell geometry, material properties, and different boundary conditions. Afterwards, the outcomes are compared with other researches. By considering the results of this article, it can be concluded that the deformation-dependent pressure assumption can entail to decrease the calculated buckling load in shells. This characteristic is studied for different examples.