• Management Science and Financial Engineering
• /
• 제11권3호
• /
• pp.19-45
• /
• 2005

Collaborative filtering is one of the most common methods that e-commerce sites and Internet information services use to personalize recommendations. Collaborative filtering has the advantage of being able to use even sparse evaluation data to predict preference scores for new products. To date, however, no in-depth investigation has been conducted on how the data sparsity effect in customers' evaluation data affects collaborative filtering-based recommendation performance. In this study, we analyzed the sparsity effect and used a hybrid method based on customers' evaluations and purchases collected from an online bookstore. Results indicated that recommendation performance decreased monotonically as sparsity increased, and that performance was more sensitive to sparsity in evaluation data rather than in purchase data. Results also indicated that the hybrid use of two different types of data (customers' evaluations and purchases) helped to improve the recommendation performance when evaluation data were highly sparse.

• 한국ITS학회 논문지
• /
• 제10권3호
• /
• pp.95-100
• /
• 2011

정보통신 기술의 급속한 성장 및 발전에 따라 유비쿼터스 네트워크 컴퓨팅 및 이용자 맞춤 서비스에 대한 관심이 증폭되고 있다. 또한 최근 스마트폰(Smartphone)을 매개체로 모바일 관련 기술들이 급속도로 발전하며 큰 각광을 받고 있다. 이러한 환경 및 인프라의 발전에 따라 최근 모바일에서 각 종 정보 및 서비스를 제공하는 다양한 응용소프트웨어들이 출시되고 있는 추세이나 그 대부분이 공급자 위주의 정보시스템으로 단순히 다량의 정보들을 불특정 다수의 이용자들에게 제공하는데 목적을 두고 있으며 이용자 개개인에 대한 맞춤화 혹은 개인화된 정보 및 서비스의 제공은 거의 이루어지지 않고 있다. 이에 따라 본 연구에서는 모바일환경에서 개인화 및 맞춤화를 위한 추천시스템을 설계 및 구현 한다. 각 종 정보필터링 기법의 장점만을 결합한 하이브리드 필터링(Hybrid Filtering)을 이용하여 추천 시스템을 구성하며 추천의 질을 향상시키기 위해 정보 필터링 단계에 앞서 사용자의 목적행위, 위치의 상황정보(Context-information)를 이용하여 추천대상 아이템의 범위를 결정함으로써 이용자 상황에 따른 효과적인 정보의 추천을 가능하도록 한다.

• Asia pacific journal of information systems
• /
• 제14권2호
• /
• pp.131-149
• /
• 2004

Collaborative filtering is one of popular techniques for personalized recommendation in e-commerce sites. An advantage of collaborative filtering is that the technique can work with sparse evaluation data to predict preference scores of new alternative contents or advertisements. There is, however, no in-depth study about the sparsity effect of customer's evaluation data to the performance of recommendation. In this study, we investigate the sparsity effect and hybrid usages of customers' evaluation data and purchase data using an experiment result. The result of the analysis shows that the performance of recommendation decreases monotonically as the sparsity increases, and also the hybrid usage of two different types of data; customers' evaluation data and purchase data helps to increase the performance of recommendation in sparsity situation.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2003년도 춘계학술대회
• /
• pp.416-425
• /
• 2003

Recommender systems are a personalized information filtering technology to help customers find the products they would like to purchase. Collaborative filtering (CF) has been known to be the most successful recommendation technology. However its widespread use in e-commerce has exposed two research issues, sparsity and scalability. In this paper, we propose several hybrid recommender procedures based on web usage mining, clustering techniques and collaborative filtering to address these issues. Experimental evaluation of suggested procedures on real e-commerce data shows interesting relation between characteristics of procedures and diverse situations.

• Journal of Power Electronics
• /
• 제17권3호
• /
• pp.704-715
• /
• 2017

For three-phase synchronization applications, the synchronous reference frame phase-locked loop (SRF-PLL) is probably the most widely used technique due to its ease of implementation and satisfactory phase tracking performance under ideal grid conditions. However, under unbalanced and distorted grid conditions, its performance tends to worsen. To deal with this problem, a variety of filtering stages have been proposed and used in SRF-PLLs for the rejection of disturbance components at the cost of degrading the dynamic performance. In this paper, to improve dynamic performance without compromising the filtering capability, an effective hybrid filtering stage is proposed and incorporated into the inner loop of a quasi-type-1 PLL (QT1-PLL). The proposed filtering stage is a combination of a moving average filter (MAF) and a modified delay signal cancellation (DSC) operator in cascade. The time delay caused by the proposed filtering stage is smaller than that in the conventional MAF-based and DSC-based PLLs. A small-signal model of the proposed PLL is derived. The stability is analyzed and parameters design guidelines are given. The effectiveness of the proposed PLL is confirmed through experimental results.

• 한국콘텐츠학회논문지
• /
• 제11권11호
• /
• pp.1-9
• /
• 2011

차세대 IT융합기술의 발전에 따라 개인화 건강관리 서비스를 위한 인프라스트럭처가 구축되면서, 사용자의 선호도 기반 서비스의 중요성이 부각되고 있다. 만성질환자가 증가함에 따라 건강관리 서비스는 특정 질병의 치료 및 관리에서 서비스 대상자에 대한 식이영양 관리로 진화하고 있다. 본 논문에서는 건강관리를 위한 혼합 필터링을 이용한 개인화 식이영양 콘텐츠 추천을 제안하였다. 제안한 방법에서는 유사한 콘텐츠만을 제공하는 추천 특수화 경향을 개선하기 위해 협력적 필터링과 이미지 기반 필터링을 결합한 혼합 필터링을 사용한다. 이를 웹 어플리케이션으로 구축하여 논리적 타당성과 유효성을 검증하기 위해 실험적인 적용을 시도한다. 따라서 식이영양 콘텐츠를 추천하여 건강관리에 대한 만족도와 서비스의 질을 향상시켰다. 연구결과를 활용하면 시장성 증대와 고부가 가치를 창출할 수 있을 것으로 기대하며 다양한 응용분야에 활용이 가능하다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 추계종합학술대회 논문집(1)
• /
• pp.145-148
• /
• 2000

This paper presents the performance of a GHICW(Group-wise Hybrid Interference Cancellation scheme based on Wiener filtering) receiver for the multi-rate DS-CDMA system. Our scheme has a small processing delay and a simple hardware complexity compared to ordinary interference cancellation schemes by grouping users with the same date rate. The performance improvement of the low rate user is obtained by using a Wiener filter which precisely estimates the high rate users' bit.

• 대한임베디드공학회논문지
• /
• 제12권6호
• /
• pp.395-403
• /
• 2017

Image processing for object detection and identification has been studied for supply chain management application with various approaches. Among them, feature pointed detection algorithm is used to track an object or to recognize a position in automated supply chain systems and a depth image based feature point detection is recently highlighted in the application. The result of feature point detection is easily influenced by image noise. Also, the depth image has noise itself and it also affects to the accuracy of the detection results. In order to solve these problems, we propose a novel hybrid filtering mechanism for depth image based feature point detection, it shows better performance compared with conventional hybrid filtering mechanism.

• Journal of Power Electronics
• /
• 제7권3호
• /
• pp.257-264
• /
• 2007

Active power filters can be cost-effective for use in practical systems with the insertion of a few passive elements in shunt or series configuration. The resulting hybrid filters can be designed to provide dominant lower order harmonic elimination and reactive power support by the passive elements so that the burden on the active filter counterpart is reduced. In this paper, the rate reduction in the shunt active filter is estimated when it is connected in parallel with suitable passive tuned harmonic filters. The active filtering system is based on an appropriate control scheme. The simulation and the experimental results of the shunt active filter, along with the estimated value of reduction in rating, show that the hybrid shunt filtering system is quite effective in compensating for the harmonics and reactive power, in addition to being cost-effective.

• 한국빅데이터학회지
• /
• 제2권1호
• /
• pp.41-48
• /
• 2017

최근 추천 시스템의 인기와 함께 추천 시스템의 알고리즘의 성능에 대한 평가가 중요해 졌다. 본 연구는 영화 데이터에서 다양한 알고리즘 중 어떤 알고리즘의 효과적인지 판단하기 위하여 모델링과 RMSE를 통한 모델 검증을 하였다. 본 연구의 데이터는 무비렌즈의 평가 데이터 10만건을 활용하여 피어슨 상관계수를 활용한 사용자 기반 협업 필터링, 코사인 상관계수를 활용한 아이템 기반 협업 필터링 그리고 특이 값분해를 활용한 아이템 기반 협업 필터링 모델을 만들었다. 세가지 추천 모델로 평점을 예측한 결과 사용자 기반 협업 필터링보다 아이템 기반 협업 필터링의 정확도가 월등히 높은 것을 확인했고, 행렬 분해를 사용했을 때 더 정확한 추천을 할 수 있었다.