• The Journal of Korean Institute of Next Generation Computing
• /
• v.14 no.6
• /
• pp.15-29
• /
• 2018

This paper proposes an enhanced object tracking algorithm to compensate the lack of temporal information in existing particle swarm optimization based object trackers using the trajectory of the target object. The proposed scheme also enables the tracking and documentation of the location of an online updated set of distractions. Based on the trajectories information and the distraction set, a rule based approach with adaptive parameters is utilized for occlusion detection and determination of the target position. Compare to existing algorithms, the proposed approach provides more comprehensive use of available information and does not require manual adjustment of threshold values. Moreover, an effective weight adjustment function is proposed to alleviate the diversity loss and pre-mature convergence problem in particle swarm optimization. The proposed weight function ensures particles to search thoroughly in the frame before convergence to an optimum solution. In the existence of multiple objects with similar feature composition, this algorithm is tested to significantly reduce convergence to nearby distractions compared to the other existing swarm intelligence based object trackers.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.23 no.3
• /
• pp.501-514
• /
• 2013

A grouping process through the similarity comparison is required to effectively classify and respond a malicious code. When we have a use of the past similarity criteria to be used in the comparison method or properties it happens a increased problem of false negatives and false positives. Therefore, in this paper we apply to choose variety of properties to complement the problem of behavior analysis on the heuristic-based of 2nd step in malicious code auto analysis system, and we suggest a similarity comparison method applying AHP (analytic hierarchy process) for properties weights that reflect the decision-making technique. Through the similarity comparison of malicious code, configured threshold is set to the optimum point between detection rates and false positives rates. As a grouping experiment about unknown malicious it distinguishes each group made by malicious code generator. We expect to apply it as the malicious group information which includes a tracing of hacking types and the origin of malicious codes in the future.

• Journal of KIISE:Software and Applications
• /
• v.36 no.4
• /
• pp.329-334
• /
• 2009

The ontology alignment has two kinds of major problems. First, the features used for ontology alignment are usually defined by experts, but it is highly possible for some critical features to be excluded from the feature set. Second, the semantic and the structural similarities are usually computed independently, and then they are combined in an ad-hoc way where the weights are determined heuristically. This paper proposes the modified parse tree kernel (MPTK) for ontology alignment. In order to compute the similarity between entities in the ontologies, a tree is adopted as a representation of an ontology. After transforming an ontology into a set of trees, their similarity is computed using MPTK without explicit enumeration of features. In computing the similarity between trees, the approximate string matching is adopted to naturally reflect not only the structural information but also the semantic information. According to a series of experiments with a standard data set, the kernel method outperforms other structural similarities such as GMO. In addition, the proposed method shows the state-of-the-art performance in the ontology alignment.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.175-177
• /
• 1999

본 논문에서는 수사 정보와 문장간 유사도를 이용하여 문서의 수사 구조 트리를 생성하는 방법을 제안하였다. 말뭉치에서 찾아낸 수사 정보를 종류별로 분류하고, 이를 사용해서 문서 내의 수사 정보를 추출해서 가능한 모든 구조를 생성한다. 다음으로 문장간의 유사도를 사용해서 가중치가 가장 높은 하나의 구조를 선택한다. 생성된 수사 구조를 사용하여 문서를 요약할 수 있는데, 수사 정보는 언어적 특성을 이용하는 것이므로 모데인에 독립적인 요약 시스템을 만들 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.172-174
• /
• 1999

본 논문에서는 카운셀링을 원하는 사용자가 카운셀러와 전자메일을 통해 상담을 원할 때 사용자의 상담 내용에 근거하여 유사한 사례를 검색해 주는 시스템을 제안한다. 제안방법은 문서의 계층적 클러스터링과 용어 적합성 피드백을 상담 사례 검색 시스템에 적용시켜, 상담사례에 나타나는 단어의 출현 빈도와 유사도를 통해 트리 구조를 형성하고, 이 트리 구조를 통한 하향 탐색을 수행한다. 하향 탐색을 하는 도중 노드의 매칭함수의 값이 서로 유사하여 노드 선택이 어려울 경우, 사용자에게 질의를 통해 용어를 제시하고, 사용자의 피드백을 통해 입력된 사연 내용의 가중치를 개선하여 내용에 가장 부합되는 문서를 탐색한다.

• Proceedings of the KIEE Conference
• /
• 2009.07a
• /
• pp.189_190
• /
• 2009

본 논문에서는 선로사고 발생시 주변 지역의 유사한 파급효과가 나타나는 선로들을 분류하기 위해 사고시 주변모선 전압을 바탕으로 일정한 기준에 따라 그룹으로 나누고 각 그룹별 가중치와 모선수를 이용해 파급효과가 얼마나 유사한지를 나타내는 지수인 SI를 구하고 이를 이용해 각 선로의 파급영향을 분석해 보았다.

• Proceedings of the Korean Society for Information Management Conference
• /
• 2000.08a
• /
• pp.37-40
• /
• 2000

KNN(K-Neatest Neighbors)을 사용한 문서의 자동분류에서는 새로운 입력문서에 범주를 할당하기 위해 K개의 유사문서로부터 범주별 문서의 분류빈도나 유사도를 이용한다. 본 연구에서는 KNN 기법에서 보편적으로 사용되는 범주 할당 방법을 응용하여 K개 유사문서 중 최상위 및 상위 M개 문서에 가중치를 부여하는 방법들을 고안하였고 K값의 변화에 따른 이들의 성능을 비교해 보았다.

• The KIPS Transactions:PartD
• /
• v.11D no.3
• /
• pp.717-730
• /
• 2004

Recently. many researches on the personalization of a web-site have been actively made. The web personalization predicts the sets of the most interesting URLs for each user through data mining approaches such as clustering techniques. Most existing methods using clustering techniques represented the web transactions as bit vectors that represent whether users visit a certain WRL or not to cluster web transactions. The similarity of the web transactions was decided according to the match degree of bit vectors. However, since the existing methods consider only whether users visit a certain URL or not, users' interestingness on the URL is excluded from clustering web transactions. That is, it is possible that the web transactions with different visit proposes or inclinations are classified into the same group. In this paper. we propose an enhanced transaction modeling with interestingness weight to solve such problems and a new similarity measuring method that exploits the proposed transaction modeling. It is shown through performance evaluation that our similarity measuring method improves the accuracy of the web transaction clustering over the existing method.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.5
• /
• pp.61-69
• /
• 2014

Collaborative filtering recommendation creates similar item subset or similar user subset based on user preference about items and predict user preference to particular item by using them. Thus, if preference matrix has low density, reliability of recommendation will be sharply decreased. To solve these problems we suggest Hybrid Preference Prediction Technique Using Weighting based Data Reliability. Preference prediction is carried out by creating similar item subset and similar user subset and predicting user preference by each subset and merging each predictive value by weighting point applying model condition. According to this technique, we can increase accuracy of user preference prediction and implement recommendation system which can provide highly reliable recommendation when density of preference matrix is low. Efficiency of this system is verified by Mean Absolute Error. Proposed technique shows average 21.7% improvement than Hao Ji's technique when preference matrix sparsity is more than 84% through experiment.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.279-284
• /
• 2009

Speaker verification determines whether the claimed speaker is accepted based on the score of the test utterance. In recent years, methods based on Gaussian mixture models and universal background model have been the dominant approaches for text-independent speaker verification. These speaker verification systems based on these methods provide very good performance under laboratory conditions. However, in real situations, the performance of speaker verification system is degraded dramatically. For overcoming this performance degradation, the feature recombination method was proposed, but this method had a drawback that whole sub-band feature vectors are used to compute the likelihood scores. To deal with this drawback, a modified feature recombination method which can use each sub-band likelihood score independently was proposed in our previous research. In this paper, we propose a sub-band weighting method based on sub-band signal-to-noise ratio which is combined with previously proposed modified feature recombination. This proposed method reduces errors by 28% compared with the conventional feature recombination method.