• 한국정보과학회논문지:소프트웨어및응용
• /
• 제31권2호
• /
• pp.153-163
• /
• 2004

대부분의 추천 시스템에서는 개인의 선호 정보를 바탕으로 한 내용-기반 추천 기법과 다른 사람들로부터의 추천을 기반으로 한 사회적 추천 기법을 사용한다. 이들 두 기법은 각각 장단점을 갖고 있으며, 서로 경쟁 관계에 있다기보다 상호 보완적인 성격을 갖고 있다. 이에 두 기법의 적절한 조합이 전체 추천 시스템의 질을 결정하는 관건이 된다. 본 논문에서는 사용자 개인마다 각 기법에 대한 만족도와 의존도가 다름을 밝히고, 이러한 각 개인의 경향에 따라 여러 추천 기법의 결과를 개인별로 조합해 주는 기법을 제안하였다. 각 개인의 경향을 나타내는 척도로 충성도, 다양도, 전문가도 둥의 척도를 정의하여 사용하였으며, 이 원리에 의해 동작하는 조합 엔진의 결과는 최고 40%, 평균 23%의 coverage 개선 효과를 나타내었다.

• ETRI Journal
• /
• 제28권4호
• /
• pp.509-512
• /
• 2006

In a real environment, sound recordings are commonly distorted by channel and background noise, and the performance of audio identification is mainly degraded by them. Recently, Philips introduced a robust and efficient audio fingerprinting scheme applying a differential (high-pass filtering) to the frequency-time sequence of the perceptual filter-bank energies. In practice, however, the robustness of the audio fingerprinting scheme is still important in a real environment. In this letter, we introduce alternatives to the frequency-temporal filtering combination for an extension method of Philips' audio fingerprinting scheme to achieve robustness to channel and background noise under the conditions of a real situation. Our experimental results show that the proposed filtering combination improves noise robustness in audio identification.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.2045-2050
• /
• 2004

An optimal combination of arbitrary number correlated estimates is derived. In particular, for two estimates this combination represents the well-known Millman and Bar-Shalom-Campo formulae for uncorrelated and correlated estimation errors, respectively. This new result is applied to the various estimation problems as least-squares estimation, Kalman filtering, and adaptive filtering. The new approximate adaptive filter with a parallel structure is proposed. It is shown that this filter is very effective for multisensor systems containing different types of sensors. Examples demonstrating the accuracy of the proposed filter are given.

• 한국컴퓨터정보학회논문지
• /
• 제23권12호
• /
• pp.219-226
• /
• 2018

Collaborative filtering has been most popular approach to recommend items in online recommender systems. However, collaborative filtering is known to suffer from data sparsity problem. As a simple way to overcome this problem in literature, Jaccard index has been adopted to combine with the existing similarity measures. We analyze performance of such combination in various data environments. We also find optimal weights of factors in the combination using a genetic algorithm to formulate a similarity measure. Furthermore, optimal weights are searched for each user independently, in order to reflect each user's different rating behavior. Performance of the resulting personalized similarity measure is examined using two datasets with different data characteristics. It presents overall superiority to previous measures in terms of recommendation and prediction qualities regardless of the characteristics of the data environment.

• International Journal of Control, Automation, and Systems
• /
• 제4권4호
• /
• pp.456-465
• /
• 2006

The problem of recursive filtering far linear discrete-time systems with uncertainties is considered. A new suboptimal filtering algorithm is herein proposed. It is based on the fusion formula, which represents an optimal mean-square linear combination of local Kalman estimates with weights depending on cross-covariances between local filtering errors. In contrast to the optimal weights, the suboptimal weights do not depend on current measurements, and thus the proposed algorithm can easily be implemented in real-time. High accuracy and efficiency of the suboptimal filtering algorithm are demonstrated on the following examples: damper harmonic oscillator motion and vehicle motion constrained to a plane.

• Tribology and Lubricants
• /
• 제24권6호
• /
• pp.320-325
• /
• 2008

The recent industrial application requires technical methods to get the cutting fluid droplet surfaces in particular from the viewpoint of topography and micro texture. To characterize the surface topography of droplet, the combination of the confocal laser scanning microscope (CLSM) and wavelet filtering is well suited for obtaining the droplet geometry encountered in tribological research. This technique indicates a better agreement in obtaining an appropriate droplet surface obtained by the CLSM over a detail range of surface accuracy (resolution: $2{\mu}m$). And the results allow an excellent accuracy in a measurement of a droplet surface. The combination of extended focal depth measurement configured and multi-scale wavelet filtering has proven that it can construct a droplet surface in a successive and accurate way. A multi-scale approach of wavelet filtering was developed based on the decomposition and reconstruction of droplet surface by 2D wavelet transform using db9 (a mother wavelet of daubechies). Also this technique can be extended to characterize the quantification of droplet properties and other field in a wide range of scales. Finally this method is verified to be a better droplet surface modeling in a micro scale arising in a mist machining.

• 한국멀티미디어학회논문지
• /
• 제9권2호
• /
• pp.255-263
• /
• 2006

스팸 메일의 특성을 분석해 보면 스팸 메일 발송 프로그램이 메일 헤더에 기록된 주소와 송신자 및 수신자 메일 주소가 일치하지 않는 경우가 빈번하게 발견된다. 또한, 스팸 메일과 정상적인 메일을 비교-분석해 보면 제목만 살펴봐도 스팸 메일인지 여부를 쉽게 판별할 수가 있다. 본 논문에서는 이와 같은 스팸 메일의 특성을 이용하여 스팸 메일 필터링 시스템의 성능을 향상시키는 방안으로 메일 주소 유효성 검사 및 제목과 내용을 구분하여 각각 스팸 확률을 계산하는 기법을 제안하였다. 제안한 방법의 효용성을 검증하기 위하여 단순 베이스 기법에 대해 주소 유효성 검사 및 제목과 내용 등 각 요인의 중요도에 따른 스팸 메일 필터링의 성능 향상 정도를 측정하였다. 그 결과로, 제안한 방법을 적용했을 때 재현율이 11.6%, 정확률은 2.1%의 성능 향상 효과가 있음을 확인하였으며, 스팸 메일 필터링 시스템의 성능 향상에 많은 기여를 하는 것을 알 수 있었다.

• Journal of electromagnetic engineering and science
• /
• 제14권2호
• /
• pp.79-81
• /
• 2014

In B-scan data measured by a pulse-type monostatic borehole radar, target signatures are seriously obscured by two clutters that differ in orientation and intensity. The primary clutter appears as a nearly constant time delay, which is caused by internal ringing between antenna and transceiver in the radar system. The secondary clutter occurs as an oblique time delay due to the guided borehole wave along the logging cable of the radar antenna. This issue led us to perform adaptive filtering processing for orientation-based clutter removal. This letter describes adaptive filtering processing consisting of a combination of edge detection, data rotation, and eigenimage filtering. We show that the hyperbolic signatures of a dormant air-filled tunnel target can be more distinctly enhanced by applying the proposed approach to the B-scan data, which are measured in a well-suited test site for underground tunnel detection.

• KIEE International Transaction on Systems and Control
• /
• 제2D권2호
• /
• pp.115-119
• /
• 2002

In this paper we present a generalized directional morphological filtering algorithm for the removal of impulse noise, which is based on a combination of impulse noise detection and a weighted rank-order morphological filtering technique. For salt (or pepper) noise suppression, the generalized directional opening (or closing) filtering of the input signal is selectively used. The detection of impulse noise can be done by the geometrical difference of opening and closing filtering. Simulations show that this new filter has better detail feature preservation with effective noise reduction compared to other nonlinear filtering techniques.

• 전자공학회논문지B
• /
• 제31B권2호
• /
• pp.21-27
• /
• 1994

Linear combination of weighted order statistic(LWOS) filters, which is an extension of stack filters, can represent any Boolean function(BF) or its extension. Which is called the extended BF(EBF). In this paper, we present a procedure for finding an LWOS filter of the simplest type from LWOS filters which are equivalent to a given BF or EBF. In addition, a property that is useful for implementing an LWOS filter is derived and an algorithm for LWOS filtering is presented.