• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.70-72
• /
• 2001

오늘날 대부분의 인터넷 경매 시스템은 단지 가격이라는 단일 속성만을 상품 거래의 절대 기준으로 삼고 있다. 이러한 경매 시스템은 단일 판매자와 다수의 구매자가 경매에 참여하므로 구매자에게 공정만 협상을 할 수 있는 기회를 제공하지 못한다. 이러한 문제를 해결하기 위해서 제품을 거래할 때 가격 외에도 배달시간, 보증기간 등 다중 속성을 가지고 협상을 수행할 수 있는 지능형 에이전트가 요구된다. 또한 다수의 판매자와 다수의 구매자가 협상을 진행할 수 있는 M:N 지능형 협상 에이전트도 필요하다. 본 논문에서는 Multi-Attribute Utility Theory(HAUT)를 이용하여 다중 속성을 가지고 거래를 하고 각각의 속성별로 거래 우선순위 가중치를 두어 협상하는 경매시스템을 제안한다. 그리고 다수의 판매자와 구매자가 협상을 하여 실제 협상과 차이가 많은 온라인 경매 협상 방법의 문제에 대한 해결방안도 제시하였다. 또한 사용자의 만족도를 높이기 위해 각 에이전트에게 최적의 상대 에이전트와 거래할 수 있는 예약시스템을 구현하였다. 제안한 모델은 약 85%의 거래 성사비율과 약 80%의 거래만족도를 보여준다.

• The Journal of Information Technology and Database
• /
• v.4 no.2
• /
• pp.3-19
• /
• 1998

관계형 데이터베이스 성능을 향상시키는데 중요한 요소 중의 하나는 트랜잭션을 처리하기 위해 데이터를 디스크에서 주 기억장치로 옮기는데 필요한 디스크 액세스(access) 횟수이다. 본 연구는 관계형 데이터베이스에서 트랜잭션을 처리할 때, 릴레이션(relation)을 속성의 중복할당을 허용하여 분할하고, 디스크에 단편(fragment)으로 저장하므로써 필요한 단편만을 액세스하여 디스크의 액세스 횟수를 줄이는 방법을 연구하였다. 본 연구에서는 속성의 중복할당을 허용하여 디스크의 액세스 횟수를 최소화시킬 수 있는 수직분할문제에 수리모형을 조회, 갱신트랙잭션을 모두 고려하여 0-1 정수계획법으로 개발하였다. 또한 모형에 대한 최적해법으로 분지한계법을 제안하였으며, 분지한계법으로 큰 문제를 푸는데는 많은 시간이 소요되므로 계산량을 줄일 수 있는 초기처리방법과 비용계산방법을 제안하였다. 속성의 중복을 허용하여 구한 해가 중복을 고려하지 않은 경우의 해보다 디스크 액세스횟수가 감소한 것으로 나타났으며, 갱신트랜？션의 수가 증가함에 따라 중복되는 속성의 수가 감소하는 결과를 나타내었다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.462-466
• /
• 2008

In this paper, we propose a Data Dependant Swarm Intelligence Routing Algorithm(DSRA) based on "ant colony optimization" to improve routing performance in Mobile Ad hoc Network(MANET). DSRA generates a different routing path depending on data's characteristics: Realtime and Non-Realtime. DSRA achieves a reduced delay for Realtime data and an enhanced network lifetime from a decentralized path selection for Non-Realtime data. We demonstrate these results by an experimental study comparing with AODV, DSR and AntHocNet.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10a
• /
• pp.68-70
• /
• 1998

본 논문에서는 클록들을 주기적으로 동기화하는 분산 실시간 시스템에서 주어진 태스크의 시간 제약(timing constraint)을 변환시는 구가지 기법을 제안한다. 전형적인 이산 클록 동기화(discrete clock synchronization)알고리즘은 클록의 값을 순간적으로 보정(correct)하여 클록의 시간이 불연속적으로 진행학 한다. 이러한 시간상의 불연속성은 태스크의 시작제한시간(release time)이나 종료시한(deadline)과 같은 이벤트를 잃어버리거나 다시 발생시키는 오류를 범하게 한다. 클록 시간의 불연속성을 피하기 위해 일반적으로 연속 클록 동기화(continuous clock synchronization) 기법이제안되었지만 소프트웨어적으로 구현되기에는 많은 오버헤드를 유발시키는 문제점이 있다. 이에 따라 연속 클록 동기화는 PLL (Phase-Locked Loop)을 이용한 별도의 하드웨어를 사용하는 것이 보통이다. 본 논문에서는 연속 클록 동기화 기법을 사용하는 대신, 태스크의 시간 제약을 동적으로 변환시키는 DCT (Dynamic Constraint Transformation) 기법을 제안하였다. DCT는 소프트웨어 으로 구현이 가능하여 새로운 하드웨어를 필요로 하지 않으며, 이를 통해 기존의 이산적으로 동기화된 시스템에서 클록 시간의 불연속성에 의한 문제점들을 해결할 수 있다. 또 다른 문제점으로서, 클록의 물리적인 특성으로 인해 동기화된 클록들이 상한된(bounded from the above)오차(skew)를 갖는다는 것이다. 이러한 오차는 지역 클록(local clock)에 대해 만족될 수 있는 임의의 실기간 제약 조건이 전역 클록(global clock)에 대해서는 만족되지 않을 수 있음을 의미한다. 본 논문에서는 이를 위해 먼저 두 가지의 스케줄링 가능성, 지역적 스케줄링 가능서(local schedulability)과 전역적 스케줄링 가능성(global schedulability)을 정의하고, 실시간 제약을 정적으로 변환시키는 SCT (Static Constraint Transformation)기법을 제안하였다. SCT를 통해 지역적으로 스케줄링 가능한 태스크는 전역적으로 스케줄링이 가능하므로, 단지 지역적 스케줄링 가능성만을 검사하면서 스케줄링 문제를 해결할 수 있도록 하였다.

• Journal of KIISE:Software and Applications
• /
• v.31 no.12
• /
• pp.1684-1690
• /
• 2004

This paper presents a method for classifier selection that uses distribution information of the training samples in a small region surrounding a sample. The conventional DCS-LA(Dynamic Classifier Selection - Local Accuracy) selects a classifier dynamically by comparing the local accuracy of each classifier at the test time, which inevitably requires long classification time. On the other hand, in the proposed approach, the best classifier in a local region is stored in the FSA(Feature Space Attribute) table during the training time, and the test is done by just referring to the table. Therefore, this approach enables fast classification because classification is not needed during test. Two feature space attributes are used entropy and density of k training samples around each sample. Each sample in the feature space is mapped into a point in the attribute space made by two attributes. The attribute space is divided into regular rectangular cells in which the local accuracy of each classifier is appended. The cells with associated local accuracy comprise the FSA table. During test, when a test sample is applied, the cell to which the test sample belongs is determined first by calculating the two attributes, and then, the most accurate classifier is chosen from the FSA table. To show the effectiveness of the proposed algorithm, it is compared with the conventional DCS -LA using the Elena database. The experiments show that the accuracy of the proposed algorithm is almost same as DCS-LA, but the classification time is about four times faster than that.

• Journal of KIISE
• /
• v.44 no.4
• /
• pp.344-351
• /
• 2017

An attribute based signature system is a cryptographic system where users produce signatures based on some predicate of attributes, using keys issued by one or more attribute authorities. If a private key is leaked during signature generation, the signature can be forged. Therefore, signing operation computations should be performed using secure hardware, which is called tamper resistant hardware in this paper. However, since tamper resistant hardware does not provide high performance, it cannot perform many operations requiring attribute based signatures in a short time frame. This paper proposes a new attribute based signature system using high performance general hardware and low performance tamper resistant hardware. The proposed signature scheme consists of two signature schemes within a existing attribute based signature scheme and a digital signature scheme. In the proposed scheme, although the attribute based signature is performed in insecure environments, the digital signature scheme using tamper resistant hardware guarantees the security of the signature scheme. The proposed scheme improves the performance by 11 times compared to the traditional attribute based signature scheme on a system using only tamper resistant hardware.

• The Journal of the Korea Contents Association
• /
• v.10 no.9
• /
• pp.201-211
• /
• 2010

The aim of this study is to investigate the method to enhance user cognition for "STM information(Short-term Memorable Information)" that is relatively accessible to information in a short period of time in information design types. What stands out from this study is the design attributes and expression method of information in a broad perspective. By 4 visualization attributes of function variable; 'Operations', 'Events', 'Methods' and 'Use cases', STM information should be satisfied by the attribute of 'Understandable' and 'Accessibility' from the point of view of visual representation and by the attribute of 'Errorless' and 'Timeliness' from the point of view of user operation. As the expression method of each perspectives, I suggested "Attribution theory", "Cognitive model", "Maximization of Proactivity", "Minimization of surplus information" and "Using dual-code" in the point of view of visual representation, and "Context effect", "Using memory code" and "Two methods of information scanning" in the point of view of user operation. I assured that above-mentioned methods are efficient and cognitive pattern of user for STM information is found out by survey and interview.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.108-110
• /
• 2015

In this paper, we proposed the relay node selection scheme using property of node such as velocity, direction in Delay Tolerant Networks. the existing selection scheme is caused the problem increasing the transmission delay and packet loss, if select the relay node for different mobile with destination. To overcome this problem, we proposed the relay node selection scheme using the property of mobile node. the proposed scheme represents and shares the property of mobile node. The proposed algorithm assumed the sketchy position of node from mobile node delivering property of destination. In addition, the propose algorithm recognizes and analyzes the context of mobile node to provide the relay node transferring the data efficiently. The simulation result provides the better result in terms of transmission delay and packet delivery ratio by selecting transmission by relay node according property of node.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.867-876
• /
• 2023

In this study, the attributes of the NHPP software development cost model applying the Gamma family distribution (Erlang, Log-Logistic, Rayleigh) were newly analyzed, and after comparing with the Goel-Okumoto basic model to verify the properties of the model, the optimal model was also presented based on this. To analyze software reliability, failure time data that occurred randomly during system operation was used, and the calculation of the parameters was solved using the maximum likelihood estimation. As a result of comprehensive evaluation through various attribute analysis (mean value function, development cost, optimal release time), it was confirmed that the Rayleigh model had the best performance. Through this study, the attributes of the software development cost model applying the Gamma family distribution, which has no previous research case, were newly identified. Also, basic design data could also be presented so that developers can efficiently utilize this research data at an early stage.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.42 no.9
• /
• pp.1499-1512
• /
• 2013

The purpose of this study was to compare the coffee shop visits of Koreans and Mongolians, and to determine the quality attributes that should be managed by Importance-Performance Analysis (IPA). The survey was conducted in Seoul and the Gyeonggi Province of Korea, and at Ulaanbaatar in Mongolia from April to May 2012. The questionnaire was distributed to 380 Koreans and 380 Mongolians, with 253 and 250 responses from the Koreans and Mongolians, respectively, used for statistical analyses. From the results, Koreans visited coffee shops more frequently than Mongolians, with both groups mainly visiting a coffee shop with friends. Koreans also spent more time in a coffee shop than Mongolians. In addition, they generally used a coffee shop, regardless of time. In terms of coffee preference, Koreans preferred Americano and Mongolians preferred Espresso. The most frequently stated purpose of Koreans for visiting a coffee shop was to rest, while Mongolians typically visited to drink coffee. The general price range respondents spent on coffee was less than 4~8 thousand won for the Koreans and 2~4 thousand won for the Mongolians. Both Koreans and Mongolians obtained information about coffee shops from recommendations. According to the IPA results of 20 quality attributes of coffee shops, the selection attributes with high importance but low satisfaction were quality, price, and kindness for Koreans, but none of the attributes was found for Mongolians.