• Journal of Broadcast Engineering
• /
• v.23 no.3
• /
• pp.403-408
• /
• 2018

Generally, a large data set is required for learning of deep learning. However, since it is not easy to create large data sets, there are a lot of techniques that make small data sets larger through data expansion such as rotation, flipping, and filtering. However, these simple techniques have limitation on extendibility because they are difficult to escape from the features already possessed. In order to solve this problem, we propose a method to acquire new image data by using existing data. This is done by retrieving and acquiring similar images using existing image data as a query of the content-based image retrieval (CBIR). Finally, we compare the performance of the base model with the model using CBIR.

• The Journal of the Korea Contents Association
• /
• v.7 no.4
• /
• pp.66-73
• /
• 2007

Internet is a part of our common life and tremendous information is cumulated. In these trends, the personalization becomes a very important technology which could find exact information to present users. Previous personalized services use content based filtering which is able to recommend by analyzing the content and collaborative filtering which is able to recommend contents according to preference of users group. But, collaborative filtering needs the evaluation of some amount of data. Also, It cannot reflect all data of users because it recommends items based on data of some users who have similar inclination. Therefore, we need a new recommendation method which can recommend prefer items without preference data of users. In this paper, we proposed the apparel coordination system using personalized preference on the semantic web. This paper provides the results which this system can reduce the searching time and advance the customer satisfaction measurement according to user's feedback to system.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.99-105
• /
• 2021

In addition to the OTT video production service represented by Nexflix and YouTube, a personalized recommendation system for content with artificial intelligence has become common. YouTube's personalized recommendation service system consists of two neural networks, one neural network consisting of a recommendation candidate generation model and the other consisting of a ranking network. Netflix's video recommendation system consists of two data classification systems, divided into content-based filtering and collaborative filtering. As the online platform-led content production is activated by the Corona Pandemic, the field of virtual influencers using artificial intelligence is emerging. Virtual influencers are produced with GAN (Generative Adversarial Networks) artificial intelligence, and are unsupervised learning algorithms in which two opposing systems compete with each other. This study also researched the possibility of developing AI platform based on individual recommendation and virtual influencer (metabus) as a core content of OTT in the future.

• Journal of Korea Multimedia Society
• /
• v.13 no.3
• /
• pp.467-474
• /
• 2010

The market of Internet-based fashion/coordination shopping malls have been growing rapidly year by year. In accordance with this growth, Internet fashion shopping malls are also making a lot of efforts to increase their revenue by displaying new fashion products on a high spot or by having professional models wear them to make them more attractive to the customers. If online shopping malls have the functionality of automatically calculating the matching degree of lower and upper clothes, it could play a role of off-line shop assistants and provide a more convenient way of purchasing fashion products for customers. In this paper, we present a learning system adopting the content-based filtering method for online shopping malls, which automatically calculates the matching degree of lower and upper clothes and recommends the most well-matched pair.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2000.11a
• /
• pp.37-40
• /
• 2000

전자상거래 분야에서 급속도로 증가하고 있는 정보들 중에서 사용자가 자신의 기호에 맞는 정보들만 선택하기 위하여 각 정보를 일일이 검토하기는 어려운 일이다. 이를 보완하기 위해 자동화된 정보 여과 기술이 사용되는데 대표적인 방법들로 내용 기반 여과(content-based filtering)와 협력적 여과(collaborative filtering)가 있다. 이 중 협력적 여과 기술은 높은 질(quality)을 가지는 정보들을 여과할 수 있지만 해당 정보의 속성은 고려하지 않는다는 난점을 가지는데, 본 논문에서는 이를 보완하여 정보의 대표 속성을 중심으로 선호도 예측을 수행하는 개선된 협력적 여과 방법을 제안한다. 그리고 기존의 협력적 여과 기술들과 예측의 정확성에 대하여 성능 비교 실험을 수행함으로써 제안한 방법의 타당성을 제시한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.2
• /
• pp.242-247
• /
• 2009

Recently an intelligent system is developed for the service what users want not a passive system which just answered user's request. This intelligent system is used for personalized recommendation system and representative techniques are content-based and collaborative filtering. In this study, we propose a prediction system which is based on the techniques of recommendation system using a collaborative filtering and a fuzzy system to solve the collaborative filtering problems. In order to verify the prediction system, we used the data that is user's rating about movies. We predicted the user's rating using this data. The accuracy of this prediction system is determined by computing the RMSE(root mean square error) of the system's prediction against the actual rating about the each movie and is compared with the existing system. Thus, this prediction system can be applied to base technology of recommendation system and also recommendation of multimedia such as music and books.

• Genomics & Informatics
• /
• v.7 no.2
• /
• pp.136-140
• /
• 2009

Genome-wide association (GWA) studies may benefit from the inclusion of imputed SNPs into their dataset. Due to its predictive nature, the imputation process is typically not perfect. Thus, it would be desirable to develop a scheme for filtering out the imputed SNPs by maximizing the concordance with the observed genotypes. We report such a scheme, which is based on the combination of several parameters that are calculated by PLINK, a popular GWA analysis software program. We imputed the genotypes of 8,842 Korean individuals, based on approximately 2 million SNP genotypes of the CHB＋JPT panel in the International HapMap Project Phase II data, complementing the 352k SNPs in the original Affymetrix 5.0 dataset. A total of 333,418 SNPs were found in both datasets, with a median concordance rate of 98.7%. The concordance rates were calculated at different ranges of parameters, such as the number of proxy SNPs (NPRX), the fraction of successfully imputed individuals (IMPUTED), and the information content (INFO). The poor concordance that was observed at the lower values of the parameters allowed us to develop an optimal combination of the cutoffs (IMPUTED${\geq}$0.9 and INFO${\geq}$0.9). A total of 1,026,596 SNPs passed the cutoff, of which 94,364 were found in both datasets and had 99.4% median concordance. This study illustrates a conservative scheme for filtering imputed SNPs that would be useful in GWA studies.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2399-2413
• /
• 2021

A recommendation system is an information filter tool, which uses the ratings and reviews of users to generate a personalized recommendation service for users. However, the cold-start problem of users and items is still a major research hotspot on service recommendations. To address this challenge, this paper proposes a high-efficient hybrid recommendation system based on Fuzzy C-Means (FCM) clustering and supervised learning models. The proposed recommendation method includes two aspects: on the one hand, FCM clustering technique has been applied to the item-based collaborative filtering framework to solve the cold start problem; on the other hand, the content information is integrated into the collaborative filtering. The algorithm constructs the user and item membership degree feature vector, and adopts the data representation form of the scoring matrix to the supervised learning algorithm, as well as by combining the subjective membership degree feature vector and the objective membership degree feature vector in a linear combination, the prediction accuracy is significantly improved on the public datasets with different sparsity. The efficiency of the proposed system is illustrated by conducting several experiments on MovieLens dataset.

• Asia pacific journal of information systems
• /
• v.15 no.3
• /
• pp.223-241
• /
• 2005

In spite of the rapid growth of mobile multimedia contents market, most of the customers experience inconvenience, lengthy search processes and frustration in searching for the specific multimedia contents they want. These difficulties are attributable to the current mobile Internet service method based on inefficient sequential search. To overcome these difficulties, this paper proposes a MOBIIe COntents Recommender System for Movie(MOBICORS-Movie), which is designed to reduce customers' search efforts in finding desired movies on the mobile Internet. MOBICORS-Movie consists of three agents: CF(Collaborative Filtering), CBIR(Content-Based Information Retrieval) and RF(Relevance Feedback). These agents collaborate each other to support a customer in finding a desired movie by generating personalized recommendations of movies. To verify the performance of MOBICORS-Movie, the simulation-based experiments were conducted. The results from this experiments show that MOBICORS-Movie significantly reduces the customer's search effort and can be a realistic solution for movie recommendation in the mobile Internet environment.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.991-1000
• /
• 2020

We propose a content-based image retrieval system that uses a combination of completed local binary pattern (CLBP) and color autocorrelogram. CLBP features are extracted on a multi-resolution multi-direction filtered domain of value component. Color autocorrelogram features are extracted in two dimensions of hue and saturation components. Experiment results revealed that the proposed method yields a lot of improvement when compared with the methods that use partial features employed in the proposed method. It is also superior to the conventional CLBP, the color autocorrelogram using R, G, and B components, and the multichannel decoded local binary pattern which is one of the latest methods.