• Journal of KIISE:Databases
• /
• v.30 no.5
• /
• pp.507-520
• /
• 2003

In this paper, an automatic construction method of theme melody index for large music database and an associative content-based music retrieval mechanism in which the constructed theme melody index is mainly used to improve the users' response time are proposed. First, the system automatically extracted the theme melody from a music file by the graphical clustering algorithm based on the similarities between motifs of the music. To place an extracted theme melody into the metric space of M-tree, we chose the average length variation and the average pitch variation of the theme melody as the major features. Moreover, we added the pitch signature and length signature which summarize the pitch variation pattern and the length variation pattern of a theme melody, respectively, to increase the precision of retrieval results. We also proposed the associative content-based music retrieval mechanism in which the k-nearest neighborhood searching and the range searching algorithms of M-tree are used to select the similar melodies to user's query melody from the theme melody index. To improve the users' satisfaction, the proposed retrieval mechanism includes ranking and user's relevance feedback functions. Also, we implemented the proposed mechanisms as the essential components of content-based music retrieval systems to verify the usefulness.

• Journal of KIISE:Databases
• /
• v.31 no.2
• /
• pp.174-182
• /
• 2004

In order to produce information the author usually reference other authors' work. A citation index leads users to papers by citations. Citations lead the user to desired information. In this paper, KSCI(Korea Science Citation Index) which defines the relationships between citing documents and cited documents has been constructed. KSCI System is to solve problems for recursive retrieval in ISI's SCI(Science Citation Index) Path Encoding Indexing technique was used to solve the problems. From the analysis of data, this system has efficiency about 8.98% in the aspect of data storage. In the aspect of retrieval, there was efficiency between citing documents and cited documents, especially there was over 40% of efficiency in the retrieval of cited documents. It is concluded that suggested KSCI system will provide efficient storage and retrieval system.

• Analytical Science and Technology
• /
• v.31 no.5
• /
• pp.195-200
• /
• 2018

Researchers have used the visual criteria to compare the effectiveness of fingerprint enhancement techniques. However, since the visual criteria is subjective, bias can occur. Therefore, many researchers have tried to make the criteria objective using the contrast index (CI). However, there is a constraint that the CI was only validated on inked fingerprints or color-change fingerprints. In this study, we compared enhancement of fluorescence fingerprints on colored papers using the CI. The CI values differ according to the color of the background, reagent, and light source, even though the same standard fingerprints were used. Through the comparison of values, we could compare the enhancement of fluorescence fingerprints according to these factors. When visual evaluation criteria were used, the same peak score was obtained regardless of the reagent and light source. However, we could get a more detailed score comparison in this study using the CI. Therefore, it is confirmed that for comparison of enhancement of fluorescence fingerprints, the CI can be used as a basis for determining which light source and reagent are appropriate according to the background.

• Journal of the Korean GEO-environmental Society
• /
• v.16 no.6
• /
• pp.23-30
• /
• 2015

This study analyzed degree of disturbance, using specimens by laboratory test with large block specimens and piston samplers collected from the Korea's two typical soft ground districts: South coast Gwangyang and Yangsan. To assess the characteristics of compression index of laboratory test incurred by disturbance, the compression index of laboratory test was compared with the back analysis compression index resulting from the analysis of the measured settlement. The analysis of specimen disturbance of the laboratory test results with the piston specimens of the two districts found that the qualities of most specimens were poor and the settlement predicted by the laboratory test compression index was underestimated. The analysis of test material taken from nearby areas proved that the disturbance degrees of large block specimens were lower than that of the piston specimens. The hyperbolic method, Hoshino method, Asaoka method, and ${\sqrt{S}}$ method, all of which are predictive methods using measured settlement, were employed to reach a conclusion that reliabilities of each predictive method except predictive material of a few points were the same. To compensate the disturbance effects on compression index of the piston specimens, we suggested a new modification formula that estimates compression index of piston specimens, using Schmertmann's corrected compression index, and back analysis compression index from the analysis of predictive settlement.

• Proceedings of the Safety Management and Science Conference
• /
• 1999.11a
• /
• pp.209-216
• /
• 1999

정보 검색 분야의 문서 분류에 기계 학습 기법을 적용할 때 발생하는 가장 큰 문제는 문서를 패턴으로 표현할 때, 하나의 패턴이 가지는 특징의 수가 기계 학습 기법에서 처리할 수 있는 범위를 넘어서는 것이다. 이러한 문제를 해결하기 위하여 특징 선택 기법은 패턴을 구성하고 있는 특징 중에서 실제 문서 분류에 많은 영향을 주는 특징만을 선택하여, 기계 학습 기법에서 쉽게 처리할 수 있을 정도의 패턴을 구성하게 한다. 본 논문에서는 이러한 특징 선택 기법 중에서 IG(Information Gain), Gini index, Relief-F, DF(Document Frequency)를 비교하였다. 실험 결과 문서들에 포함된 모든 고유 단어를 특징의 길이로 하여 패턴을 구성했을 때보다 특징 선택 기법을 적용하여 고유 단어 중 일부를 특징으로 패턴을 구성할 때 기계학습에서 더 향상된 분류 성능을 보였다

• Proceedings of the Korea Institutes of Information Security and Cryptology Conference
• /
• 1994.11a
• /
• pp.166-179
• /
• 1994

Diffie-Hellman 의 공개 키 암호 프로토콜이 제안된 이후 이산 대수 문제의 어려움이 프로토콜의 안전도와 깊이 연관되었다. 유한체를 이용한 암호 기법을 ElGamal 이 세웠으나, Index-Calculus 알고리듬에 의해 유한체위 에서 이산 대수 문제가 subexponential 알고리듬이되 어 ElGamal 기법의 안전도가 약해졌다. Nonsupersingular타원 곡선을 선택하여 유한체대신 ElGamal 암호 기법에 적용하면 안전한 암호 시스템을 설계할 수 있다. 이 논문에서는 콤퓨터 구현시 용이한 nonsupersingular 타원 곡선을 선택하는 방법, 유한체위에서의 연산, 평문을 타원 곡선의 원소로 임베드(Imbedding) 하는 방법 등 타원 곡선을 암호시스템에 적응하기 어려운 점들에 대한 해결 방법을 소개하고, 실제로 콤퓨터로 구현하여 그 실행 결과와 ElGamal 기법을 개선한 Schnorr 기법을 실행한 결과를 밝혔다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.54-56
• /
• 2021

데이터베이스 백업은 데이터베이스 데이터를 보존 및 복구하기 위해 사용되는 필수 기법이다. 이를 위해 데이터베이스 관리 시스템(Database Management System)에서는 백업 기능을 수행하는 응용을 제공한다. 하지만 이러한 백업 응용은 기존 HDD 의 특성을 고려하여 설계 및 구현되어있기 때문에 최근 주로 사용되고 있는 저장장치인 SSD 에서는 SSD 의 성능을 최대한 이끌어내지 못하고 있다. 예를 들어, 기존 백업 응용에서는 단일 스레드가 백업을 위한 데이터를 순차적으로 접근하고 풀 스캔 방식으로 백업을 수행한다. 따라서 본 연구에서는 SSD 의 병렬적 특성을 활용하여 백업 응용의 성능을 극대화시키는 것을 목표로 한다. 이를 위해 본 논문은 병렬 인덱스 스캔 기법을 제시한다. 제안한 기법은 기존 백업 기법과 달리 멀티 스레드를 활용하여 인덱스 스캔을 수행하고 병렬적으로 데이터에 접근하여 백업을 수행한다. 해당 기법은 기존 기법에 비해 최대 2.5 배 성능을 향상시켰다.

• Journal of Korea Water Resources Association
• /
• v.39 no.9 s.170
• /
• pp.801-812
• /
• 2006

In this study, Combined Drought Index which can monitor the drought severity and intensity has been developed using PDSI, SPI and MSWSI. To verify the accuracy and applicability of combined drought index, Drought map of Korea using the combined drought index has compared with past drought event. Drought map using the combined drought index shows good accordance with past drought event and accurate quantitative drought monitoring results. Also the drought outlook technique has been developed using the weather forecast data of Korea Meteorological Administration (KMA). Drought outlook technique of this study can be used effectively as a primitive stage tool for real time drought forecast. As a result of this study, Integrated drought monitoring system has been developed which has capabilities of producing and generating the drought monitoring map and drought outlook map as well as various kinds of drought related information.

• The KIPS Transactions:PartD
• /
• v.10D no.3
• /
• pp.547-558
• /
• 2003

From the viewpoint of that music melody has the similar features to time-series data, music melody is transformed to a time-series data with normalization and corrections and the similarity between melodies is defined as the Euclidean distance between the transformed time-series data. Then, based the similarity between melodies of a music object, melodies are clustered and the representative of each cluster is extracted as one of theme melodies for the music. To construct the theme melody index, a theme melody is represented as a point of the multidimensional metric space of M-tree. For retrieval of user's query melody, the query melody is also transformed into a time-series data by the same way of indexing phase. To retrieve the similar melodies to the query melody given by user from the theme melody index the range query search algorithm is used. By the implementation of the prototype system using the proposed theme melody index we show the effectiveness of the proposed methods.

• Journal of Wetlands Research
• /
• v.20 no.3
• /
• pp.201-209
• /
• 2018

In this study, the water quality index was calculated using the water quality monitoring data of the major tributaries of the Nakdong River and long-term trend analysis was performed to identify the tributaries requiring priority management. We used a Real-Time Water Quality Index method implemented by the Ministry of Environment. Linear regression as a parametric method and Mann-Kendall Test and Sen Slope Test as a nonparametric method were applied for the trend analysis. The water quality index of major tributaries except for Migeon2 and Seokyo2 was in the range below Fair grade and there were no significant trends for the rest of the sites except Bukan, Chennae, Hogye, Yongdeok. Therefore, in order to improve the water quality of the main stream, management of the tributaries should be preceded.