• Communications for Statistical Applications and Methods
• /
• 제2권2호
• /
• pp.184-197
• /
• 1995

An objective and one-step detection procedure of multiple outliers is suggested by using the gaps of the order statistics. The detection procedure can be used as a routine outlier detection method of a statistical analysis computer program. The procedure is applied to some examples including the data selected by Kitagawa.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2003년도 Proceeding
• /
• pp.141-148
• /
• 2003

Customer retention is a common concern for many industries and a critical issue for the survival in today's greatly compressed marketplace. Current customer retention models only focus on detection of potential defectors based on the likelihood of defection by using demographic and customer profile information. In this paper, we propose a dynamic procedure for defection detection and prevention using past and current customer behavior by utilizing SOM and Markov chain. The basic idea originates from the observation that a customer has a tendency to change his behavior (i.e. trim-out his usage volumes) before his eventual withdrawal. This gradual pulling out process offers the company the opportunity to detect the defection signals. With this approach, we have two significant benefits compared with existing defection detection studies. First, our procedure can predict when the potential defectors could withdraw and this feature helps to give marketing managers ample lead-time for preparing defection prevention plans. The second benefit is that our approach can provide a procedure for not only defection detection but also defection prevention, which could suggest the desirable behavior state for the next period so as to lower the likelihood of defection. We applied our dynamic procedure for defection detection and prevention to the online gaming industry. Our suggested procedure could predict potential defectors without deterioration of prediction accuracy compared to that of the MLP neural network and DT.

• 대한의용생체공학회:의공학회지
• /
• 제8권2호
• /
• pp.205-214
• /
• 1987

In this paper, a mathmatical model for the purpose of QRS detection is considered in the case of the occurence of nonoverlappjng pulse-shaped waveforms corrupted with white noise. The number of waveform, the arrival times, amplitudes, and widths of QRS complexes are regarded as random variables. The joint MAP estimation of all the unknown Quantities consists of linear filtering followed by an optimization procedure. Because the optimization procedure is time-consuming, this procedure is modified so that a threshold test is obtained.

• 해양환경안전학회지
• /
• 제20권5호
• /
• pp.564-570
• /
• 2014

본 논문에서는 컬러정보와 깊이정보를 사용하여 얼굴을 검출하고 추적한 후 항해사의 졸음을 탐지하는 방법을 제안한다. 이 방법은 얼굴검출 과정과 얼굴추적 과정으로 구성된다. 얼굴검출 과정에서는 기존의 방법 중 가장 좋은 성능을 보이는 Adaboost 알고리즘을 사용하며, Adaboost로 입력되는 영상의 영역을 사람이 존재하는 영역으로만 제한하여 얼굴을 검출한다. 얼굴검출 과정에서 얼굴이 검출되면 그것을 템플릿으로 하여 얼굴추적 과정이 수행된다. 제안한 방법의 성능을 평가하기 위하여 실험영상을 이용하여 실험을 수행하였다. 실험결과 제안한 졸음탐지 방법은 기존의 방법에 비해 약 23 %의 수행시간을 보였으며, 또한 졸음탐지 방법은 추적 시간과 추적 정확도에 있어서 상보적인 관계를 가지며, 특별한 경우를 제외한 모든 경우에서 약 1 %의 낮은 추적오차율을 보였다.

• 한국재난정보학회 논문집
• /
• 제10권3호
• /
• pp.452-457
• /
• 2014

화학재난 발생시 현장대응 요원들은 사건 원인물질의 성상과 잔류오염 농도를 신속 정확하게 파악해야 한다. 또한 화학재난 현장에서의 적절한 대응절차 진행을 위해서는 화학물질의 성상과 오염농도 확인은 필수적이다. 이를 위해 현장에서 사용하는 각 장비의 특징을 알아보고자 한다. 현장대응장비는 모든 화학물질을 확인할 수 없으며, 각 장비별로 물질탐지에 제한적이다. 장비별 물질탐지 범위와 상호보완성을 고려해야 한다. 본 연구에서는 현장 활용장비인 간이탐지 킷과 검지관식 탐지장비, 전자식 탐지장비의 신속한 현장 활용을 위한 대응절차를 마련하여 현장대응에 도움을 주고자 한다.

• Food Science and Biotechnology
• /
• 제18권2호
• /
• pp.407-412
• /
• 2009

In this study, a rapid and efficient concentrating procedure that can be used for detecting viruses in vegetables was developed. The Sabin strain of poliovirus type 1 was used to evaluate the efficiency of virus recovery. The procedure included: (a) elution with 0.25 M threonine-0.3 M NaCl pH 9.5; (b) polyethylene glycol (PEG) 8000 precipitation; (c) chloroform extraction; (d) 2$^{nd}$ PEG precipitation; (f) RNA extraction; (g) reverse transcription-polymerase chain reaction (RT-PCR) combined with semi-nested PCR. The overall recoveries by elution/concentration were 29.0% from cabbage and 13.7% from lettuce. The whole procedure usually takes 18 hr. The overall detection sensitivity was 100 RT-PCR units of genogroup II norovirus (GII NoV)/25 g cabbage and 100 RT-PCR units of GII NoV/10 g lettuce. The virus detecting method developed in this study should facilitate the detection of low levels of NoV in cabbage and lettuce.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1987년도 한국자동제어학술회의논문집; 한국과학기술대학, 충남; 16-17 Oct. 1987
• /
• pp.709-712
• /
• 1987

In this paper, a mathematical model for the purpose of QRS detection is considered in the case of the occurrence of nonoverlapping pulse-shaped waveforms corrupted with white noise. The number of waveforms, the arrival times, amplitudes, and widths of QRS complexes are regarded as random variables. The joint MAP estimation of all the unknown quantities consists of linear filtering followed by an optimization procedure. Because of time-consuming, the optimization procedure is modified so that a threshold test is obtained. The model formulation with nonoverlapping waveforms leads to a standard procedure covering a segment before as well as after an accepted event. Adaptivity of the detector is gained by utilizing past signal properties in determining threshold for QRS detection.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1999년도 추계학술대회 논문집 학회본부 A
• /
• pp.279-281
• /
• 1999

This paper addresses the issues of the parameter error detection and identification in power system. The parameter error identification is carried out as part of the state estimation procedure. The weighted least absolute value(WLAV) estimation method is used for this procedure. The standard formulation of the state estimation problem is modified to include the effects of the parameter errors as well. A two step procedure for the detection and identification of faulted parameters is proposed. Supporting examples are given using IEEE 14 bus system.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.306-312
• /
• 1996

The performance of the failure detection algorithm may be greatly influenced by the model uncertainty. It is very important to design a robust failure detection system to the model uncertainty. In this paper, a design procedure to generate failure detection algorithm is proposed. The design procedure suggested is based on the concept of the‘threshold selector［1］’. The H$\infty$ control algorithm is used to derive a threshold selector which is robust to the model uncertainty, The threshold selector derived can be used to develop a failure detection system together with the weighted cumulative sum algorithm［3］. Computer simulation study showed that the failure detection system designed for an ISA(Integrated Servo Actuator) system by using the proposed method is robust to the model uncertainty.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2002년도 추계정기학술대회
• /
• pp.445-453
• /
• 2002

Past and current customer behavior is the best predicator of future customer behavior. This paper introduces a procedure on personalized defection detection and prevention for an online game site. The basic idea for our defection detection and prevention is adopted from the observation that potential defectors have a tendency to take a couple of months or weeks to gradually change their behavior (i.e. trim-out their usage volume) before their eventual withdrawal. For this purpose, we suggest a SOM (Self-Organizing Map) based procedure to determine the possible states of customer behavior from past behavior data. Based on this representation of the state of behavior, potential defectors are detected by comparing their monitored trajectories of behavior states with frequent and confident trajectories of past defectors. The key feature of this study includes a defection prevention procedure which recommends the desirable behavior state for the ext period so as to lower the likelihood of defection. The defection prevention procedure can be used to design a marketing campaign on an individual basis because it provides desirable behavior patterns for the next period. The experiments demonstrate that our approach is effective for defection prevention and efficient for defection detection because it predicts potential defectors without deterioration of prediction accuracy compared to that of the MLP (Multi-Layer Perceptron) neural network.