http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
소프트웨어 분야 취업 결정 요인에 대한 XAI 모델 적용 연구 : 일반대학교와 전문대학 졸업자를 중심으로
권준희,김성림 (사)디지털산업정보학회 2024 디지털산업정보학회논문지 Vol.20 No.1
The purpose of this study is to explain employment factors in the software field. For it, the Graduates Occupational Mobility Survey by the Korea employment information service is used. This paper proposes employment models in the software field using machine learning. Then, it explains employment factors of the models using explainable artificial intelligence. The models focus on both university graduates and vocational college graduates. Our works explain and interpret both black box model and glass box model. The SHAP and EBM explanation are used to interpret black box model and glass box model, respectively. The results describes that positive employment impact factors are major, vocational education and training, employment preparation setting semester, and intern experience in the employment models. This study provides a job preparation guide to universitiy and vocational college students that want to work in software field.
이커머스 환경에서 구매와 공유 행동을 이용한 기기 중심 개인화 상품 정보 추천 기법
권준희 (사)디지털산업정보학회 2022 디지털산업정보학회논문지 Vol.18 No.4
Personalized recommendation technology is one of the most important technologies in electronic commerce environment. It helps users overcome information overload by suggesting information that match user's interests. In e-commerce environment, both mobile device users and smart device users have risen dramatically. It creates new challenges. Our method suggests product information that match user's device interests beyond only user's interests. We propose a device-centered personalized recommendation method. Our method uses both purchase and share behavior for user's devices interests. Moreover, it considers data type preference for each device. This paper presents a new recommendation method and algorithm. Then, an e-commerce scenario with a computer, a smartphone and an AI-speaker are described. The scenario shows our work is better than previous researches
사물인터넷에서 참여 기기를 고려한 개인화 정보 검색 기법
권준희 (사)디지털산업정보학회 2020 디지털산업정보학회논문지 Vol.16 No.1
Internet of Things is growing rapidly. As it evolves, the amount of data is increasing significantly. It requires a new personalized information retrieval method. Internet of Things is defined as uniquely identifiable interoperable connected object. The first definition of Internet of Things was from Things oriented perspective. However, previous studies about personalized information retrieval method do not consider Things. To meet user’s individual needs, previous studies concentrate on only human, not Things. In this paper, we propose a personalized information retrieval method considering participating device in Internet of Things. It provides personalized information using data type preference for each device. Moreover, it provides personalized results by integrating data type preference for set of devices. This paper describes a new personalized retrieval method and algorithm. It consists of five steps. Then, it presents four scenarios using proposed method. The scenarios show our work is more effective and efficient than existing one.
퍼베이시브 컴퓨팅 환경에서 소셜네트워크를 이용한 프로액티브 친구 추천 기법
권준희 (사)디지털산업정보학회 2013 디지털산업정보학회논문지 Vol.9 No.1
Pervasive computing and social network are good resources in recommendation method. Collaborative filtering is one of the most popular recommendation methods, but it has some limitations such as rating sparsity. Moreover, it does not consider social network in pervasive computing environment. We propose an effective proactive friend recommendation method using social network and contexts in pervasive computing environment. In collaborative filtering method, users need to rate sufficient number of items. However, many users don’t rate items sufficiently, because the rating information must be manually input into system. We solve the rating sparsity problem in the collaboration filtering method by using contexts. Our method considers both a static and a dynamic friendship using contexts and social network. It makes more effective recommendation. This paper describes a new friend recommendation method and then presents a music friend scenario. Our work will help e-commerce recommendation system using collaborative filtering and friend recommendation applications in social network services.
수질오염총량관리제도 적용을 위한 도시유역의 하수발생량 분석
권준희,박인혁,하성룡 한국습지학회 2009 한국습지학회지 Vol.11 No.1
기존의 하천 수질을 규제하던 농도규제방식은 오염부하의 양적증가를 통제할 수 없어 수질개선에 한계성을 지니고 있었다. 수질오염총량관리제도는 폐수 중 오염물질의 총량을 규제하여 환경기준을 달성할 수 있는 허용부하량 이내로 배출 오염물질의 총량을 할당, 규제할 수 있다. 수질오염총량관리제도의 하수발생량 산출시 실측한 값을 사용하지 않고 오염원별 발생원단위를 곱하여 산출하여 실측한 data를 이용한 산출값과 차이가 있다. 수질오염총량관리에 의한 산출결과를 보면 건기시 관거유입량 26,460.9㎥/d, 관거이송유량 17,778.6㎥/d로 실측한 data를 이용하여 산출한 관거유입량 17,106.1㎥/d, 관거이송유량 19,033.9㎥/d로 차이가 났으며, 우기시의 경우 수질오염총량관리에 의한 관거유입량 49,512.2㎥/d, 관거이송유량 18,628.7㎥/d로 실측한 data를 이용하여 산출한 관거유입량 30,918.2㎥/d, 관거이송유량 19,700.7㎥/d로 차이가 났다. 오염 부하량의 기초값인 하수발생량이 실측한 하수발생량과 차이를 보여 효율적인 제도 수행에 문제점이 있는 것으로 판단된다. The regulation of emission concentration for stream water qualities doesn't control quantitative increase on pollution loads, it has limits for improvement of water qualities. Total water pollution load management system(TMDL) can control the total amount of pollutant in waste water which is allowed to assign and control the total discharged pollutant loads in a permissible level. When it comes to generated wastewater value of TMDL system, there is difference between calculated value based on individual pollutant unit load and observed value. Calculated sewer inflow, calculated sewer outflow, measured sewer inflow, and measured sewer outflow at dry season are 26,460.9㎥/d, 17,778.6㎥/d, 17,106.1㎥/d and 19,033.9㎥/d respectively, Calculated sewer inflow, calculated sewer outflow, measured sewer inflow, and measured sewer outflow at rainy season are 49,512.2㎥/d, 18,628.7㎥/d, 30,918.2㎥/d,19,700.7㎥/d respectively. This result presents the necessity to acquire the precise observed data to fulfill the efficient TMDL system.