• Educational Technology International
• /
• 제6권1호
• /
• pp.65-79
• /
• 2005

In a widely distributed environment, many occasions arise when people need to filter informationwith email clients. The existing information agents such as Maxims and Message Assistant have capabilities of filtering email messages either by an autonomous agent or by user-defined rules. FlexMA, a variation of FAMES (Flexible Asynchronous Messaging System) is proposed as an information filtering agent. Agents in our system can be scaled up to adapt user's various demands by controlling messages delivered among heterogeneous email clients. Several functionalities are split into each agent in terms of component configuration with the addition of multiple agents'cooperation and negotiation. User-defined rules are collected and executed by these agents in a semi-autonomous manner. This paper demonstrates how this design is feasible in a flexible message system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.604-625
• /
• 2018

Nowadays, SMS spam has been overflowing in many countries. In fact, the standards of filtering SMS spam are different from country to country. However, the current technologies and researches about SMS spam filtering all focus on dividing SMS message into two classes: legitimate and illegitimate. It does not conform to the actual situation and need. Furthermore, they are facing several difficulties, such as: (1) High quality and large-scale SMS spam corpus is very scarce, fine categorized SMS spam corpus is even none at all. This seriously handicaps the researchers' studies. (2) The limited length of SMS messages lead to lack of enough features. These factors seriously degrade the performance of the traditional classifiers (such as SVM, K-NN, and Bayes). In this paper, we present a new fine categorized SMS spam corpus which is unique and the largest one as far as we know. In addition, we propose a classifier, which is based on the probability topic model. The classifier can alleviate feature sparse problem in the task of SMS spam filtering. Moreover, we compare the approach with three typical classifiers on the new SMS spam corpus. The experimental results show that the proposed approach is more effective for the task of SMS spam filtering.

• 한국정보과학회 언어공학연구회:학술대회논문집(한글 및 한국어 정보처리)
• /
• 한국정보과학회언어공학연구회 2009년도 제21회 한글 및 한국어 정보처리 학술대회
• /
• pp.141-144
• /
• 2009

인터넷에 게시된 글(블로그, 포털 사이트, 게시판 등)에 대한 댓글들 중에서 중복, 광고 등의 스팸성 댓글을 추적하고 이를 필터링 하는 시스템을 개발.

• 한국멀티미디어학회논문지
• /
• 제10권11호
• /
• pp.1530-1537
• /
• 2007

본 논문은 마이크로폰을 통해 실시간으로 습득된 음성으로부터 사람의 음성 감성상태를 평상, 기쁨, 슬픔, 화남 등 4가지로 구별할 수 있는 ARM 플랫폼 기반의 음성 감성인식 시스템 구현에 관한 것이다. 일반적으로 마이크로폰으로 수신된 음성은 화자 주변의 환경 잡음과 마이크로폰의 시스템 특성 때문에 입력 음성 신호가 왜곡되고 이로 인해 시스템의 성능이 저하된다. 본 논문에서는 이러한 잡음 영향을 최소화하기 위해 비교적 단순한 구조와 적은 연산 량을 가진 이동평균(MA, Moving Average) 필터를 입력 음성의 특징벡터 열에 적용하였다. 또한, 효율적으로 감성 특징벡터를 최적화할 수 있는 SFS(Sequential Forward Selection)기법을 적용해 제안 시스템의 성능을 최적화하였으며 감성 패턴 분류기로는 SVM(Support Vector Machine)을 사용하였다. 실험 결과 제안 감성인식 시스템은 모의실험에서 약 65%, ARM 플랫폼에서 약 62%의 인식률을 보였다.

• 한국음향학회지
• /
• 제25권6호
• /
• pp.269-276
• /
• 2006

본 논문은 이동전화 (Cellular phone)를 통해 실시간으로 습득된 음성으로부터 사람의 감성 상태를 평상 혹은 화남으로 인식할 수 있는 음성 감성인식 시스템을 제안하였다. 일반적으로 이동전화를 통해 수신된 음성은 화자의 환경 잡음과 네트워크 잡음을 포함하고 있어 음성 신호의 감성특정을 왜곡하게 되고 이로 인해 인식 시스템에 심각한 성능저하를 초래하게 된다. 본 논문에서는 이러한 잡음 영향을 최소화하기 위해 비교적 단순한 구조와 적은 연산량을 가진 MA (Moving Average) 필터를 감성 특정벡터에 적용해서 잡음에 의한 시스템 성능저하를 최소화하였다. 또한 특정벡터를 최적화할 수 있는 SFS (Sequential Forward Selection) 기법을 사용해서 제안 감성인식 시스템의 성능을 한층 더 안 정화시켰으며 감성 패턴 분류기로는 k-NN과 SVM을 비교하였다. 실험 결과 제안 시스템은 이동통신 잡음 환경에서 약 86.5%의 높은 인식률을 달성할 수 있어 향후 고객 센터 (Call-center) 등에 유용하게 사용될 수 있을 것으로 기대된다.

• Journal of Information Processing Systems
• /
• 제14권4호
• /
• pp.941-947
• /
• 2018

Since the amplitudes of spin echo train in nuclear magnetic resonance logging (NMRL) are small and the signal to noise ratio (SNR) is also very low, this paper puts forward an improved de-noising algorithm based on wavelet transformation. The steps of this improved algorithm are designed and realized based on the characteristics of spin echo train in NMRL. To test this improved de-noising algorithm, a 32 points forward model of big porosity is build, the signal of spin echo sequence with adjustable SNR are generated by this forward model in an experiment, then the median filtering, wavelet hard threshold de-noising, wavelet soft threshold de-noising and the improved de-noising algorithm are compared to de-noising these signals, the filtering effects of these four algorithms are analyzed while the SNR and the root mean square error (RMSE) are also calculated out. The results of this experiment show that the improved de-noising algorithm can improve SNR from 10 to 27.57, which is very useful to enhance signal and de-nosing noise for spin echo train in NMRL.

• 한국통신학회논문지
• /
• 제38A권1호
• /
• pp.96-107
• /
• 2013

본 논문에서는 초광대역 통신시스템의 확산코드 정합필터링 방식이 서로 다른 다중 접속 방식에 따라 다중사용자 간섭신호의 전력 파라미터에 어떠한 영향을 미치는가에 대하여 개시한다. 보다 구체적으로, 다중 접속 시스템에 사용되는 확산코드 정합필터의 길이가 다중사용자 간섭신호의 평균전력, 최대전력, 그리고 최대전력 대 평균전력 비율에 미치는 영향을 분석하였다. 간단화된 환경에 대하여 이론적 분석을 수행하였고, 모의실험을 통하여 이를 검증하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.86-89
• /
• 1992

One of the most difficult problems in depth control for underwater vehicle is the effect of seaway disturbance. One component of the seaway forces is of large magnitude with a relatively narrow-band, first order component. The other component is generaly of somewhat smaller magnitude, second order component. Since the magnitude of the first order component is generally much greater than the compensating force that can be generating by the planes, it is undesirable for the controller to generate a control command. In this paper, we disigned adaptive notch filtering system using filter bank structure. Energies of each band-passed signal are obtained by MA(Moving Average) method and compared to produce center frequency. By adapting this parameter to notch filter, 1st order seaway disturbance can be removed, which lead to the improvement of automatic depth control system.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2002년도 추계학술대회 논문집
• /
• pp.80-84
• /
• 2002

In this study, researchers developed the est algorithm for artificial defects in the semic packages and performed to it by pattern recogn technology. For this purpose, this algorithm was I that researcher made software with matlab. The so consists of some procedures including ultrasonic acquistion, equalization filtering, self-organizing backpropagation neural network. self-organizing ma backpropagation neural network are belong to metho neural networks. And the pattern recognition tech has applied to classify three kinds of detective pa semiconductor packages. that is, crack, delaminat normal. According to the results, it was found estimative algorithm was provided the recognition r 75.7%( for crack) and 83.4%( for delamination) 87.2 % ( for normal).

• 한국정보처리학회논문지
• /
• 제6권5호
• /
• pp.1263-1271
• /
• 1999

This paper designs the performance analysis layer for the various performance analysis of parallel programs using event expressions are similar to the normal program language to analyze the events which display a dynamic state exchange of a program. The event expressions suggest operations for overloading and functions which are needed in performance analysis, such as a filtering operation, data format translation functions, performance analysis, static functions, and etc. By using the event expressions, the programmer can modify the event trace data to analyze the performance and analyze more easily and variously than the pre-developed tools.