Earticle

As deep learning technology in natural language and visual processing has rapidly developed, collaborative filtering-based recommendation systems using deep learning technology are being actively introduced in the recommendation field. In this study, OCF-GAN, a hybrid collaborative filtering model using GAN, was proposed to solve the one-class and cold-start problems, and its usefulness was verified through performance evaluation. OCF-GAN based on conditional GAN consists of a generator that generates a pattern similar to the actual user preference pattern and a discriminator that tries to distinguish the actual preference pattern from the generated preference pattern. When the training is completed, user preference vectors are generated based on the actual distribution of preferred items. In addition, the cold-start problem was solved by using a hybrid collaborative filtering recommendation method that additionally utilizes user and item profiles. As a result of the performance evaluation, it was found that the performance of the OCF-GAN with additional information was superior in all indicators of the Top 5 and Top 20 recommendations compared to the existing GAN-based recommender. This phenomenon was more clearly revealed in experiments with cold-start users and items.

현재 추천 시스템은 실제 사용자가 선호하는 항목을 추천하는지, 아니면 단순히 관심 정도의 항목을 추천하는지 알 수 없다는 문제와 사용자들이 매우 적어 적합한 항목을 추천할 수 없는 데이터 희소성 문제, 새로운 사용자들이 유입됨 에 따라 사용자들이 만족하는 항목을 추천하기 위해 시스템의 성능이 저하되는 Cold-Start 문제 등이 발생한다. 본 연구에 서는 이러한 문제점을 해결하기 위해 불확실한 상황이나 문제들을 반영할 수 있는 퍼지-계층적 분석(Fuzzy-Analytic Hierarchy Process)과 주어진 항목들을 비슷한 항목들끼리 모으는 데이터 군집화 알고리즘을 활용하여 사용자들에게 만족 할 수 있는 영화를 추천하기 위한 시스템을 구현하였다. 61명을 대상으로 영화 선호도에 대한 설문 조사를 실시한 데이터를 본 시스템에 적용한 결과 Fuzzy-AHP 기법을 통해서 데이터 희소성 문제를 해소할 수 있었으며, 또한 데이터 군집화 알고 리즘을 통해 새로운 사용자들이 유입되어도 사용자에게 적합한 항목이 추천되었음을 확인할 수 있었다. 향후 노이즈 데이 터나 아웃라이어(Outlier) 데이터를 걸러낼 수 있는 밀도 기반 클러스터링에 대한 연구가 필요할 것으로 생각된다.

The current recommendation systems have problems including the difficulty of figuring out whether they recommend items that actual users have preference for or have simple interest in, the scarcity of data to recommend proper items due to the extremely small number of users, and the cold-start issue of the dropping system performance to recommend items that can satisfy users according to the influx of new users. In an effort to solve these problems, this study implemented a movie recommendation system to ensure user satisfaction by using the Fuzzy-Analytic Hierarchy Process, which can reflect uncertain situations and problems, and the data partition algorithm to group similar items among the given ones. The data of a survey on movie preference with 61 users was applied to the system, and the results show that it solved the data scarcity problem based on the Fuzzy-AHP and recommended items fit for a user with the data partition algorithm even with the influx of new users. It is thought that research on the density-based clustering will be needed to filter out future noise data or outlier data.

Information asymmetry in emerging platforms with two-sided markets creates barriers for interactions, especially for products lacking established reputation, termed the reputational cold-start problem. To address the problem, this study explores the roles of platform-initiated contents (PICs). Specifically, this paper examines how inspector-based textual PIC and video-mediated PIC influence consumer demand and subsequent reviews in the accommodation-sharing context. We employ econometric analyses with propensity score weighting and two-stage residual inclusion to address potential endogeneity concerns. We find that (i) both PICs as quality signals enhance the booking rate, and (ii) the positive impact of video-mediated PIC is more pronounced for properties facing the reputational cold-start problem, a nuanced effect not observed for inspector-based PIC. Furthermore, we demonstrate that inspector-based PIC increases subsequent reviews for properties with the cold-start problem while video-mediated PIC has no significant effects on subsequent reviews for those properties. Contribution and practical implications for addressing the cold-start problem in emerging platforms are discussed.

On cold start operation of an SI engine, a catalyst shows poor performance before it reaches activation temperature. Therefore, fast warmup of the catalyst is very crucial to reduce harmful emissions. In this study, an appropriate control strategy is investigated to increase exhaust gas temperature through changes of spark timing. Combustion stability is also considered at the same time. Exhaust gas temperature and pressure of combustion chamber are measured to investigate the effects of spark timings on cold start and idle performance. Experiments showed that retarded spark timing promotes the combustion at the end of expansion stroke and increases exhaust gas temperature during cold start.

On cold start operation of an SI engine, a catalyst shows poor performance before it reaches activation temperature. Therefore, fast warmup of the catalyst is very crucial to reduce harmful emissions. In this study, an appropriate control strategy is investigated to increase exhaust gas temperature through changes of spark timing and exhaust valve timing. Combustion stability is also considered at the same time. Experiments showed that retarded spark timing promotes the combustion at the end of expansion stroke and increases exhaust gas temperature during cold start.An advance of exhaust valve timing decreases residual gas in cylinders due to decrease of valve overlap period. It helps improvement combustion stability by virtue of reduced residual gas. A control strategy of proper valve timing and spark timing is suggested in order to achieve fast light-off of the catalyst and stable operation of the engine in a cold start and idle operation

With the advent of serverless computing, cloud customers no longer needed to maintain and manage server environments directly. Instead, cloud service providers took on that role, managing and maintaining the server environment according to customer requests, a concept known as Function as a Service (FaaS). This service demonstrated improvements in operational costs and resource utilization over traditional cloud computing, offering various advantages such as enhanced scalability. However, a delay occurred in processing and returning results to user requests, a phenomenon referred to as the cold start problem. This paper proposed the Time Warming Allocation Engine (TWAE) to improve resource management and mitigate the cold start problem in Function as a Service. The proposed engine comprised a collection module, a learning module, a classification module, and an allocation module. Additionally, it utilized a list called Pre-Warming. Through this approach, it suggested directions for improving cold start issues and resource utilization according to different time periods.

도시대기오염의 주요 배출원으로 알려진 자동차에서 배출되는 VOCs는 인체에 유해할 뿐만 아니라 대기 중에서 질소산화물(NOx)과 함에 광화학반응을 통한 오존 둥 2차 오염물질인 광화학산화물을 형성하는 전구물질로 작용하기 때문에 환경학적, 보건학적으로 매우 큰 영향을 미친다. 이러한 자동차 배출 VOCs가 오존생성에 미치는 영향에 대한 연구는 갈수록 증가하고 있는 대기중 오존 농도 심화현상의 규명과 대기질 개선을 위한 기초자료로서 매우 그 필요성이 더해가고 있다. (중략)

급증하는 자동차 수요로 인해 발생되는 배출량 증가는 자동차 운행의 특성상 대도시 대기오염의 주원인으로 작용하고 있다. 자동차 배출오염물질의 배출량 추정에 있어 국내 차량의 배출계수 개발도 시급하지만 우리 나라 자동차 운행패턴에 관한 연구도 같이 병행되어야 한다. 본 연구에서는 Cold start emission과 Evaporative VOC emission의 산정에 있어서 국내 승용차의 1회 주행거리(Trip length)를 이용하여 승용차 부문 배출량 영향정도와 계절별 배출량 변화를 추정하여 보았다. (중략)

급증하는 자동차 수요로 인해 발생되는 배출량 증가는 자동차 운행의 특성상 대도시 대기오염의 주원인으로 작용하고 있다. 자동차 배출오염물질의 배출량 추정 및 관리에 있어 연간 배출량 뿐만 아니라 월별 기온변화에 따른 배출량의 시간분포 또한 고려되어야 한다. 본 연구에서는 Hot emission에 부가되는 Cold start emission과 Evaporative VOC emission의 산정에 있어서 차종/차령에 대해서 월별 배출량을 추정하여 보았다. (중략)

The main goal of e-learning systems is just-in-time knowledge acquisition. Rule-based e-learning systems, however, suffer from the mesa effect and the cold start problem, which both result in low user acceptance. E-learning systems suffer a further drawback in rendering the implementation of a natural interface in humanoids difficult. To address these concerns, even exceptional questions of the learner must be answerable. This paper aims to propose a method that can understand the learner's verbal cues and then intelligently explore additional domains of knowledge based on crowd data sources such as Wikipedia and social media, ultimately allowing for better answers in real-time. A prototype system was implemented using the NAO platform.

Importance and usage of the recommender system increases with the increase of information. The accuracy of the system recommendation primarily depends on the data. There is a problem in recommender systems, known as cold start problem. The lack of data about new products and users causes the cold start problem, and the system will not be able to give correct recommendation. This paper deals with cold start problem by comparing product specification and the review of the resembled products. The user, who likes the resembled product of the new one has more probability of taking interest in the new product as well. However, if a user disagreed with resembled product due to some reasons which the user mentioned in the reviews. The new product overcomes that issue, so the user will greatly accept the new product. Therefore, the system needs to recommend new product to those users as well, in this way the cold start problem will get resolved.

Serverless 플랫폼은 작업이 요청되었을 때 인스턴스를 생성하였다가 작업이 끝난 후 제거되기 때문에 실제 작업을 처리할 때만 요금이 부과되어 경제적이다. 하지만 작업이 요청될 경우 인스턴스가 생성되고 작업에 필요한 리소스를 가져오는 지연시간 즉, Cold Start 가 발생한다. 특히 워크플로 실행에서 각 작업이 개별적으로 처리되기 때문에 워크플로의 각 단계마다 Cold Start 가 발생하는데, 워크플로는 사전에 정의되기 때문에 이러한 Cold Start 의 누적은 회피 될 수 있다. 본 논문은 Serverless 환경에서 워크플로를 실행할 때 누적되는 Cold Start 를 완화하기 위해 앞의 단계에서 다음 단계를 미리 준비시키는 런타임 인터페이스들에 대하여 제안한다. 이미지 프로세싱 워크플로에 제안한 런타임 인터페이스를 적용시켜 Cold Start 를 제외한 워크플로의 실행시간을 최대 21%까지 줄였다.

본 연구에서는 협업필터링의 두 가지 근본적인 문제인 신규고객 추천(cold-start recommendation)과 희박성(sparsity) 문제를 해결하고자 한다. 먼저, 사회 네트워크 분석에서 가장 많이 활용 되고 있는 세 가지 중심성 지표인 연결중심성(degree centrality), 근접중심성(closeness centrality), 매개중심성(betweenness centrality)을 결합한 다양한 중심성 지표들을 만든 후 이를 기반으로 신규고객의 잠재 이웃고객을 찾고 그 이웃고객들의 구매정보를 이용하여 신규고객에게 상품을 추천하는 새로운 방법을 제시한다. 다음으로 희박성 문제를 해결하기 위하여, 구매정보가 충분한 고객에게는 협업필터링을, 그렇지 않은 고객에게는 협업필터링 대신 제시한 신규고객 추천방법을 적용하는 하이브리드 추천 방법을 제안한다. 제시한 추천 방법의 효과성을 평가하기 위하여 국내 유명 백화점 중의 하나인 H백화점의 구매 트랜잭션 데이터를 사용하여 실험하였다. 실험결과로부터 근접중심성과 매개중심성을 결합한 지표를 신규고객 추천 시에 사용할 경우 추천 성능이 가장 우수한 것으로 판명되었으며, 제안한 하이브리드 추천 방법이 기존의 협업필터링의 성능을 상당히 개선함으로써 희박성 문제를 해결할 수 있는 새로운 대안임이 입증되었다.

Collaborative Filtering (CF) suffers from two major problems:sparsity and cold-start recommendation. This paper focuses on the cold-start problem for new customers with no purchase records and the sparsity problem for the customers with very few purchase records. For the purpose, we propose a method for the new customer recommendation by using a combined measure based on three well-used centrality measures to identify the customers who are most likely to become neighbors of the new customer. To alleviate the sparsity problem, we also propose a hybrid approach that applies our method to customers with very few purchase records and CF to the other customers with sufficient purchases. To evaluate the effectiveness of our method, we have conducted several experiments using a data set from a department store in Korea. The experiment results show that the combination of two measures makes better recommendations than not only a single measure but also the best-seller-based method and that the performance is improved when applying the hybrid approach.

Experiments were carried out to investigate the characteristics of the hydrocarbon (HC) emissions and to reduce cold start hydrocarbons in gasoline engines. An HC adsorber was, used and it coated was by Pd/Rh catalyst with zeolite on a honeycomb monolith. The HCs were efficiently trapped at temperatures below $100^{\circ}C by physical adsorption. After adsorption, they were reduced gradually by the catalytic oxidation of Pd/Rh catalysts as the adsorber temperature increased above $100^{\circ}C. Increasing amounts of methane, ethylene and n-butane were emitted as the fuel-air mixture became richer and the engine speed decreased. As the temperature of adsorber increased, high-number carbons into low-number carbons. Thus, the C4 concentration decreased significantly during the first 30 seconds, and the C2 concentration increased continuously.

소셜 미디어는 모바일 어플리케이션과 웹에서 가장 많이 사용되는 미디어 중 하나이다. Nielsen사의 보고서에 따르면 소셜 네트워크 서비스와 블로그가 온라인 사용자의 주 활동 공간으로 사용되고 있으며, 미국인 중에서 온라인 활동이 왕성한 5명의 사용자중 4명은 매일 소셜 네트워크 서비스와 블로그를 방문하고 온라인 활동 시간의 23%를 소비한다고 집계하고 있다. 미국의 인터넷 사용자들은 야후, 구글, AOL 미디어 네트워크, 트위터, 링크드인 등과 같은 소셜 네트워크 서비스중 페이스북에서 가장 많은 시간을 소비한다. 최근에는 대부분의 회사들이 자신의 특정 상품에 대하여 "페이스북 페이지(Facebook Page)"를 생성하고 상품에 대한 프로모션을 진행한다. 페이스북에서 제공되는 "좋아요" 옵션은 페이스북 페이지를 통해 자신이 관심을 가지는 상품(아이템)을 표시하고 그 상품을 지지할 수 있도록 한다. 많은 영화를 제작하는 영화 제작사들도 페이스북 페이지와 "좋아요" 옵션을 이용하여 영화 프로모션과 마케팅에 이용한다. 일반적으로 다수의 스트리밍 서비스 제공업들도 영화와 TV 프로그램을 즐기며 볼 수 있는 서비스를 사용자들에게 제공한다. 이 서비스는 일반 컴퓨터와 TV 등의 단말기에서인터넷을 통해 영화와 TV 프로그램을 즉각적으로 제공할 수 있다. 스트리밍 서비스의 선두 주자인 넷플릭스는 미국, 라틴 아메리카, 영국 그리고 북유럽 국가 등에 3천만 명 이상의 스트리밍 사용자가 가입되어 있다. 또한 넥플릭스는 다양한 장르로 구성된 수백만 개의 영화와 TV 프로그램을 보유하고 있다. 하지만 수많은 콘텐츠로 인해 사용자들은 자신이 선호하는 장르에 관련된 영화와 TV 프로그램을 찾기 위해 많은 시간을 소비해야 된다. 많은 연구자들이 이러한 사용자의 불편함을 줄이기 위해 아이템에 대한 사용자가 보지 않은 아이템에 대한 선호도를 예측하고 높은 예측값을 갖는 아이템을 사용자에게 제공하기 위한 추천 시스템을 적용하였다. 협업적 여과 방법은 추천 시스템을 구축하기 위해 가장 많이 사용되는 방법이다. 협업적 여과 시스템은 사용자들이 평가한 아이템을 기반으로 각 사용자 간의 유사도를 측정하고 목적 사용자와 유사한 성향을 가진 사용자 그룹을 결정한다. 군집된 그룹은 이웃 사용자 집단으로 불리며 이를 이용하여 특정 아이템에 대한 선호도를 예측하고, 예측 값이 높은 아이템을 목적 사용자에게 추천해 준다. 협업적 여과 방법이 적용되는 분야는 서적, 음악, 영화, 뉴스 및 비디오 등 다양하지만 논문에서는 영화에 초점을 맞춘다. 이 협업적 여과 방법이 추천 시스템 내에서 유용하게 활용되고 있지만 아직 "희박성 문제"와 "콜드 스타트 문제" 등 해결해야 할 과제가 남아있다. 희박성 문제는 아이템의 수가 증가할수록 아이템에 대한 사용자의 로그 밀도가 감소하는 것이다. 즉, 전체 아이템 수에 비해 사용자가 아이템에 대해 평가한 정보가 충분하지 않기 때문에 사용자의 성향을 파악하기 어렵고, 이로 인해 사용자가 아직 평가하지 않은 아이템에 대해서 선호도를 추측하기 어려운 것을 말한다. 이 희박성 문제가 포함된 경우 적합한 이웃 사용자 집단을 형성하는데 어려움을 겪게 되고 사용자들에게 제공되는 아이템 추천의 질이 떨어지게 된다. 콜드 스타트 문제는 시스템 내에 새로 들어온 사용자 또는 아이템으로 지금까지 한 번도 평가를 하지 않은 경우에 발생

Social media has become one of the most popular media in web and mobile application. In 2011, social networks and blogs are still the top destination of online users, according to a study from Nielsen Company. In their studies, nearly 4 in 5active users visit social network and blog. Social Networks and Blogs sites rule Americans' Internet time, accounting to 23 percent of time spent online. Facebook is the main social network that the U.S internet users spend time more than the other social network services such as Yahoo, Google, AOL Media Network, Twitter, Linked In and so on. In recent trend, most of the companies promote their products in the Facebook by creating the "Facebook Page" that refers to specific product. The "Like" option allows user to subscribed and received updates their interested on from the page. The film makers which produce a lot of films around the world also take part to market and promote their films by exploiting the advantages of using the "Facebook Page". In addition, a great number of streaming service providers allows users to subscribe their service to watch and enjoy movies and TV program. They can instantly watch movies and TV program over the internet to PCs, Macs and TVs. Netflix alone as the world's leading subscription service have more than 30 million streaming members in the United States, Latin America, the United Kingdom and the Nordics. As the matter of facts, a million of movies and TV program with different of genres are offered to the subscriber. In contrast, users need spend a lot time to find the right movies which are related to their interest genre. Recent years there are many researchers who have been propose a method to improve prediction the rating or preference that would give the most related items such as books, music or movies to the garget user or the group of users that have the same interest in the particular items. One of the most popular methods to build recommendation system is traditional Collaborative Filtering (CF). The method compute the similarity of the target user and other users, which then are cluster in the same interest on items according which items that users have been rated. The method then predicts other items from the same group of users to recommend to a group of users. Moreover, There are many items that need to study for suggesting to users such as books, music, movies, news, videos and so on. However, in this paper we only focus on movie as item to recommend to users. In addition, there are many challenges for CF task. Firstly, the "sparsity problem"; it occurs when user information preference is not enough. The recommendation accuracies result is lower compared to the neighbor who composed with a large amount of ratings. The second problem is "cold-start problem"; it occurs whenever new users or items are added into the system, which each has norating or a few rating. For instance, no personalized predictions can be made for a new user without any ratings on the record. In this research we propose a clustering method according to the users' genre interest extracted from social network service (SNS) and user's movies rating information system to solve the "cold-start problem." Our proposed method will clusters the target user together with the other users by combining the user genre interest and the rating information. It is important to realize a huge amount of interesting and useful user's information from Facebook Graph, we can extract information from the "Facebook Page" which "Like" by them. Moreover, we use the Internet Movie Database(IMDb) as the main dataset. The IMDbis online databases that consist of a large amount of information related to movies, TV programs and including actors. This dataset not only used to provide movie information in our Movie Rating Systems, but also as resources to provide movie genre information which extracted from the "Facebook Page". Formerly, the user must login with their Facebook account to login to the

In line with the Super Ultra Low Emission Vehicle (SULEV) regulation, the main idea in this study has been focused on the utilization of hydrocarbon adsorber (HCA) to adsorb the excess hydrocarbons emitted during a period of engine cold-start, As main recipes of HCA materials, many types of zeolite as well as the combination of alumina and precious metals were used, Representative physico-chemical factors of zeolite such as acidic and hydrophobic properties were characterized. The optimum recipe of HCA materials was also determined. Among the acid properties of zeolites, the Si/Al ratio was found to be the most important factor to get higher hydrocarbon adsorption capacity.

Currently, many studies on green ships are under way. Fuel cell (FC) ships are of interest as future low-emission, fuel-efficient vessels. In this paper, a hybrid power management system for an FC ship was designed. The system consists of an FC, a battery, a unidirectional DC/DC converter, a bidirectional DC/DC converter, a filter, an inverter, and a propulsion component. To design the system, we analyze electric sources and converters, and create PLECS models of hybrid power management system. Then, we check the cold start sequence and perform a simulation to understand the characteristics of the hybrid power management system in an FC ship.

In this study, a new method is suggested to improve the cold start characteristics of FCVs (Fuel Cell Vehicles). To avoid freezing, the moisture is removed from the fuel cell system by using an adsorbent as soon as an FCV is fumed off. Experiments have been performed to verify this concept and evaluate the efficiency of water removal and regeneration characteristics. In the moisture removal phase, the time required to remove the water droplets from the solenoid valve was measured for various operation conditions. In the regeneration phase, the time for the regeneration was measured and the overall performance was evaluated in detail.

Most of the pollutants from passenger cars are emitted during the cold-transient phase of the FTP-75 test. In order to reduce the exhaust emissions during the cold-transient period, it is essential to warm up the catalyst as fast as possible after the engine starts, and the Unburned Exhaust Gas Ignition (UEGI) technology was developed through our previous studies to help close-coupled catalytic converters (CCC) reach the light-off temperature within a few seconds after cold-start. The UEGI system operates by igniting the unburned exhaust mixture by glow plugs installed upstream of the catalyst. The flame generates a high amount of heat, and if the heat is concentrated on a specific area of monolith surface, then thermal crack or failure of the monolith could occur. Therefore, it is very important to monitor the temperature distribution in the CCC during the UEGI operation, so the local temperatures in the monolith were measured using thermocouples. Experimental results showed that the temperature of CCC rises faster with the UEGI technology, and the CCC reaches the light-off temperature earlier than the baseline case. Under the conditions tested, the light-off time of the baseline case was 62 seconds, compared with 33 seconds for the UEGI case. The peak temperature is well under the thermal melting condition, and temperature distribution is not so severe as to consider thermal stress. It is noted that the UEGI technology is an effective method to warm up the catalyst with a small amount of thermal stress during the cold start period.