• Industrial Engineering and Management Systems
• /
• 제10권4호
• /
• pp.272-278
• /
• 2011

Decision tree as a classification tool is being used successfully in many areas such as medical diagnosis, customer churn prediction, signal detection and so on. The main advantage of decision tree classifiers is their capability to break down a complex structure into a collection of simpler structures, thus providing a solution that is easy to interpret. Since decision tree is a top-down algorithm using a divide and conquer induction process, there is a risk of reaching a local optimal solution. This paper proposes a procedure of optimally determining thresholds of the chosen variables for a decision tree using an adaptive particle swarm optimization (APSO). The proposed algorithm consists of two phases. First, we construct a decision tree and choose the relevant variables. Second, we find the optimum thresholds simultaneously using an APSO for those selected variables. To validate the proposed algorithm, several artificial and real datasets are used. We compare our results with the original CART results and show that the proposed algorithm is promising for improving prediction accuracy.

• 한국통신학회논문지
• /
• 제22권11호
• /
• pp.2451-2458
• /
• 1997

학습 신호를 사용하지 않고 심볼간의 간섭을 제거하여 전송된 데이터를 복구하는 등화 방법을 블라인드 등화(blind equalization)라 한다. 본 논문에서는 수렴 속도와 정상상태오차를 줄이기 위하여 Sato 알고리듬과 결정-지향 알고리듬의 장점을 동시에 이용하는 다단계 결정-지향 알고리듬을 제안한다. 다단계 결정-지향 알고리듬은 초기 시작 모드에서는 수렴이 보장되는 Sato 알고리듬과 똑같이 동작을 한다. 이 후 등화기가 점점 수렴함에 따라, 블라인드 등화기에서 사용하는 양자화기의 레벨 수를 증가시켜 수렴속도를 빠르게 해준다. 양자화기의 레벨이 완전히 증가하였을 경우 다단계 결정-지향 알고리듬은 수렴 후 자승 평균 오차가 작은 결정-지향 알고리듬과 똑같이 동작하게 된다. 그러므로 다단계 결정-지향 알고리듬은 빠른 수렴 속도를 보이면서도, 정상상태에서 작은 오차값을 갖는다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2006년도 춘계종합학술대회
• /
• pp.397-401
• /
• 2006

In this paper, we proposed a high precision location estimation algorithm using an enhanced decision scheme for RTLS and analyzed its performance in point of an average estimation error distance at 2D coordinates searching area, $300m\times300m$ and LOS propagation environments. Also the performance was compared with that of conventional TDOA algorithm according to the number of available reader and received sub-blink. From the results, we confirmed that the proposed location estimation algorithm using an enhanced decision scheme was able to improve an estimation accuracy even in boundary region of searching area. Moreover, effectively reduced an error distance in entire searching area so that increased the stability of location estimation in RTLS. Therefore, we verified that the proposed algorithm provided a more higher estimation accuracy and stability than conventional TDOA.

• 한국정보처리학회논문지
• /
• 제6권11호
• /
• pp.2863-2874
• /
• 1999

Apartment auction is a system that is used for the citizens to get a house. This paper deals with the implementation of a web-based intelligent decision support system using OLAP technique and data mining technique for auction decision support. The implemented decision support system is working on a real auction database and is mainly composed of OLAP Knowledge Extractor based on data warehouse and Auction Data Miner based on data mining methodology. OLAP Knowledge Extractor extracts required knowledge and visualizes it from auction database. The OLAP technique uses fact, dimension, and hierarchies to provide the result of data analysis by menas of roll-up, drill-down, slicing, dicing, and pivoting. Auction Data Miner predicts a successful bid price by means of applying classification to auction database. The Miner is based on the lazy model-based classification algorithm and applies the concepts such as decision fields, dynamic domain information, and field weighted function to this algorithm and applies the concepts such as decision fields, dynamic domain information, and field weighted function to this algorithm to reflect the characteristics of auction database.

• Journal of the Optical Society of Korea
• /
• 제14권1호
• /
• pp.22-27
• /
• 2010

Bit error rate (BER) monitoring is the ultimate goal of performance monitoring in all digital transmission systems as well as optical fiber transmission systems. To achieve this goal, optimization of the decision threshold must also be considered because BER is dependent on the level of decision threshold. In this paper, we analyze a pseudo-error counting scheme and propose an algorithm to achieve both BER monitoring and adaptive decision threshold optimization in optical fiber transmission systems. To verify the effectiveness of the proposed algorithm, we conduct computer simulations in both Gaussian and non-Gaussian distribution cases. According to the simulation results, BER and the optimum decision threshold can be estimated with the errors of < 20% and < 10 mV, respectively, within 0.1-s processing time in > 40-Gb/s transmission systems.

• 한국인터넷방송통신학회논문지
• /
• 제11권5호
• /
• pp.27-33
• /
• 2011

디지털 통신 시스템에서 채널에서 발생되는 ISI(Inter-Symbol Interference)를 줄이기 위해 사용되는 적응 등화기의 알고리즘으로 MCMA(Constant Modulus Algorithm)가 있다. MCMA는 비교적 간단한 연산량을 갖지만 어느 정도 적당한 수렴율과 정상 상태의 MSE(mean square error)를 가지므로 본 논문에서는 이를 개선키 위해 병렬 구조를 갖는 등화 방식을 제안하며, MCMA(Modified Constant Modulus Algorithm)과 MDD(Modified Decision Directed Algorithm)으로 구성이 되어있다. 제안 방식에서는 4-QAM 좌표와 16-QAM 좌표의 관계를 이용한, 새로운 비용 함수를 정의 하고, 만약 수신된 신호에 오프셋이 발생할 때, MCMA의 성능이 저하되므로 이를 극복하기 위해서 본 논문에서는 MDD(Modified Decision Direct) 알고리즘과 결합한 병렬형 구조를 적용하므로서 기존의 MCMA보다 빠른 수렴 속도와 낮은 MSE를 갖는 개선된 성능을 얻을 수 있음을 컴퓨터 시뮬레이션으로 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권6호
• /
• pp.2925-2948
• /
• 2019

In order to improve the utilization of irrigation water resources of greenhouse tomatoes, a water-saving irrigation decision-making model based on genetic optimization T-S fuzzy neural network is proposed in this paper. The main work are as follows: Firstly, the traditional genetic algorithm is optimized by introducing the constraint operator and update operator of the Krill herd (KH) algorithm. Secondly, the weights and thresholds of T-S fuzzy neural network are optimized by using the improved genetic algorithm. Finally, on the basis of the real data set, the genetic optimization T-S fuzzy neural network is used to simulate and predict the irrigation volume for greenhouse tomatoes. The performance of the genetic algorithm improved T-S fuzzy neural network (GA-TSFNN), the traditional T-S fuzzy neural network algorithm (TSFNN), BP neural network algorithm(BPNN) and the genetic algorithm improved BP neural network algorithm (GA-BPNN) is compared by simulation. The simulation experiment results show that compared with the TSFNN, BPNN and the GA-BPNN, the error of the GA-TSFNN between the predicted value and the actual value of the irrigation volume is smaller, and the proposed method has a better prediction effect. This paper provides new ideas for the water-saving irrigation decision in greenhouse tomatoes.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2000년도 춘계공동학술대회 논문집
• /
• pp.235-238
• /
• 2000

This article presents an adaptive decision tree algorithm for dynamically reasoning machine failure cause out of real-time, large-scale machine status database. On the basis of experiment using semiconductor etching machine, it has been verified that our model outperforms previously proposed decision tree models.

• 한국정보통신학회논문지
• /
• 제8권6호
• /
• pp.1121-1125
• /
• 2004

본 논문은 H.264/AVC를 위한 고속 블록 모드 결정 알고리즘에 대해 연구한다. H.264/AVC는 단일 크기의 블록을 이용하여 움직임 추정을 수행하는 기존 동영상 부호화 방식과는 다르게 16$\times$16, 16$\times$8, 8$\times$16, 8$\times$8, 8$\times$4, 4$\times$8, 4$\times$4의 7가지 서브 블록을 이용하는 가변 블록 움직임 추정을 채택한다. 이 방식은 효율적인 움직임 추정을 가능하게 하지만 동영상 부호화의 연산량을 크게 증가시키는 원인으로 작용한다. 고속 블록 모드 결정 알고리즘은 먼저 매크로블록에 존재하는 4개의 8$\times$8 블록을 기준으로 8$\times$8 보다 큰 블록 모드와 8$\times$8 보다 작은 블록 모드로 블록 영역을 예측한다. 여기서 8$\times$8 보다 큰 블록 모드로 예측되면 각 8$\times$8 블록간의 움직임 벡터 거리를 임계값과 비교하여 8$\times$8 이상의 블록 모드로 합병한다. 이는 8$\times$8보다 큰 블록 모드의 움직임 추정을 위해 RDO를 16$\times$16, 16$\times$8, 8$\times$16 8$\times$8에 대해 모두 수행하는 것이 아니라 각각의 8$\times$8 블록에 대해서만 수행하므로 블록 모드 결정을 위한 연산량을 효율적으로 감소시킬 수 있다.

• ETRI Journal
• /
• 제32권4호
• /
• pp.577-587
• /
• 2010

The layered coding structure of scalable video coding (SVC) with adaptive inter-layer prediction causes noticeable computational complexity increments when compared to existing video coding standards. To lighten the computational complexity of SVC, we present a fast algorithm to speed up the inter-mode decision process. The proposed algorithm terminates inter-mode decision early in the enhancement layers by estimating the rate-distortion (RD) cost from the macroblocks of the base layer and the enhancement layer in temporal, spatial, and inter-layer directions. Moreover, a search range decision algorithm is also proposed in this paper to further increase the motion estimation speed by using the motion vector information from temporal, spatial, or inter-layer domains. Simulation results show that the proposed algorithm can determine the best mode and provide more efficient total coding time saving with very slight RD performance degradation for spatial and quality scalabilities.