• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.603-605
• /
• 2022

추천시스템은 개개인의 성향에 따른 맞춤화 추천이 가능하기 때문에 음악, 영상, 뉴스 등 많은 분야에서 관심을 받고 있다. 일반적인 추천시스템 모델은 블랙박스 모델이기 때문에 추천 결과에 따른 원인 도출을 할 수 없다. 하지만 XAI 의 모델은 이러한 블랙박스 모델의 단점을 해결하고자 제안되었다. 그 중 KGAT 는 Attention Score 를 기반으로 추천 결과에 따른 원인을 알 수 있다. 이와 같은 AI, XAI 등의 딥 러닝 모델에서 각각의 활성화 함수는 상황에 따라 상이한 성능을 나타낸다. 이러한 이유로 인해 데이터에 맞는 활성화 함수를 적용해보는 다양한 시도가 필요하다. 따라서 본 논문은 XAI 추천시스템 모델인 KGAT 의 성능 개선을 위해 여러 활성화 함수를 적용해보고, 실험을 통해 수정한 모델의 성능이 개선됨을 보인다.

• The Journal of Bigdata
• /
• v.8 no.2
• /
• pp.149-163
• /
• 2023

Due to the continuous growth of the E-commerce market, the volume of orders that fulfillment centers have to process has increased, and various customer requirements have increased the complexity of order processing. Along with this trend, the operational efficiency of fulfillment centers due to increased labor costs is becoming more important from a corporate management perspective. Using historical performance data as training data, this study focused on real-time box recommendations applicable to packaging areas during fulfillment center shipping. Four types of data, such as product information, order information, packaging information, and delivery information, were applied to the machine learning model through pre-processing and feature-engineering processes. As an input vector, three characteristics were used as product specification information: width, length, and height, the characteristics of the input vector were extracted through a feature engineering process that converts product information from real numbers to an integer system for each section. As a result of comparing the performance of each model, it was confirmed that when the Gradient Boosting model was applied, the prediction was performed with the highest accuracy at 95.2% when the product specification information was converted into integers in 21 sections. This study proposes a machine learning model as a way to reduce the increase in costs and inefficiency of box packaging time caused by incorrect box selection in the fulfillment center, and also proposes a feature engineering method to effectively extract the characteristics of product specification information.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.323-338
• /
• 2023

TV advertising with deep analysis of watching pattern of audiences is important to set-top box audience targeting. Applying session-based recommendation model(SBR) to internet commercial, or recommendation based on searching history of user showed its effectiveness in previous studies, but applying SBR to the TV advertising was difficult in South Korea due to data unavailabilities. Also, traditional SBR has limitations for dealing with user preferences, especially in data with user identification information. To tackle with these problems, we first obtain set-top box data from three major broadcasting companies in South Korea(SKB, KT, LGU+) through collaboration with Korea Broadcast Advertising Corporation(KOBACO), and this data contains of watching sequence of 4,847 anonymized users for 6 month respectively. Second, we develop personalized session-based recommendation model to deal with hierarchical data of user-session-item. Experiments conducted on set-top box audience dataset and two other public dataset for validation. In result, our proposed model outperformed baseline model in some criteria.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.291-294
• /
• 2021

본 논문에서는 영어 교육에 대한 사회적인 인식은 중요해지면서 다양한 방식의 영어공부와 학습모델들을 분석하고 오랫동안 가억이 가능한 학습시스템을 검토하였다. 그러나 영어의 기초가 되는 영어 단어의 공부법이 논리적인 근거 없이 강사의 인지도와 명성에 현혹되어 강사가 추천해 주는 공부 방법에 따라 잘못된 학습하는 경우가 대부분이다. 또한 사람마다 자기에게 맞는 학습법은 분명 존재할 것이다. 하지만 그 방법 외에 효과적인 다른 방법 또한 찾아 볼 수 있다. 헤르만 에빙하우스는 사람의 망각 연구한 결과 망각의 주기를 수치로 나타내었고, 라이트너는 플래시 카드를 박스에 넣어 복습을 반복하는 아날로그 장치를 만들어 복습의 효과를 연구하였다. 본 논문에서는 헤르만 에빙하우스의 망각 곡선 이론을 통해 효율적인 영어 단어 학습방법을 논리적으로 증명하고, 웹사이트를 사용한 라이트너 박스와 망각 곡선의 이론을 적용하여 망각의 주기에 따른 복습의 효과를 적용하여 결과적으로 사용자에게 접근성이 좋은 학습형 플랫폼을 제공하여 시간 대비 학습률이 좋은 시스템을 개발하고자 하였다.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.1013-1023
• /
• 2019

In content consumption environment with various live TV channels, VoD contents and web contents, recommendation service is now a necessity, not an option. Currently, various kinds of recommendation services are provided in the OTT service or the IPTV service, such as recommending popular contents or recommending related contents which similar to the content watched by the user. However, in the case of a content viewing environment through TV or IPTV which shares one TV and a TV set-top box, it is difficult to recommend proper content to a specific user because one or more usage histories are accumulated in one subscription information. To solve this problem, this paper interprets the concept of family as {user, time}, extends the existing recommendation relationship defined as {user, content} to {user, time, content} and proposes a method based on deep learning algorithm. Through the proposed method, we evaluate the recommendation performance qualitatively and quantitatively, and verify that our proposed model is improved in recommendation accuracy compared with the conventional method.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.187-190
• /
• 2015

주행 중 차선 이탈 경고 시스템은 사고 발생 예방 차원에서 매우 높은 효과가 인정되어서 차선이탈 경고 장치(LDWS) 제품들이 출시되고 있다. 본 논문은 블랙박스의 영상을 이용하여 차선 검출에 정확도를 향상하기 위한 알고리즘을 연구한 것으로 특히 차량에 장착되어 있는 블랙박스 영상을 영상 변환 없이, 실시간 소프트웨어 만 으로 처리할 수 있는 알고리즘을 연구한다. 차선인식을 위한 최적의 영상 ROI를 결정하고, 차선 인식 정확도를 향상하기 위한 전 처리 과정을 적용하고, 동영상의 연속성을 잘못된 차선인식에 대한 보정, 인식이 되지 않는 차선에 대한 후보 차선 추천 알고리즘과 시점 변환에 의한 야간, 곡선 도로에 대한 오인식율을 최소화 하는 방법을 제안한다. 도로주행의 다양한 환경에 대한 실험을 진행했으며, 각각의 방법 적용에 의한 오인식율의 감소와 많은 인식 알고리즘 적용에 의한 처리 속도 저하를 개선하기 위한 연구를 진행했으며, 본 논문은 블랙박스 영상을 이용하여 주행 차선 인식을 위한 최적 알고리즘을 제안한다.

• Journal of the Korean Society of Radiology
• /
• v.17 no.1
• /
• pp.131-139
• /
• 2023

In this study, we are purposed to find the educational effect of practical training using a 3D printed structure-based blind box on multidimensional radiographic image interpretation. The subjects were 83 (male: 49, female: 34) 2nd year radiological science students who participated in the digital medical imaging practice that was conducted for 3 years from 2020 to 2022. The learning method used 3D printing technology to print out the inside structure of the blind box designed by itself. After taking X-rays 3 times (x, y, z axis), the structure images in the blind box were analyzed for each small group. We made the 3D structure that was self-made with clay based on our 2D radiographic images. After taking X-rays of the 3D structure, it was compared whether it matches the structural image of the blind box. The educational effect for the practical training surveyed class faithfulness, radiographic image interpretation ability (attenuation concept, contrast concept, windowing concept, 3-dimensional reading ability), class satisfaction (interest, external recommendation, immersion) on a 5-point Likert scale as an anonymous student self-writing method. As a result, all evaluation items had high positive effects without significant differences between males and females. Practical education using blind boxes is a meaningful example of radiology education technology using 3D printing technology, and it is expected to be used as content to improve students' problem-solving skills and increase satisfaction with major subjects.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.23-43
• /
• 2022

In order to apply AI (Artificial Intelligence) in various industries, interest in algorithm selection is increasing. Algorithm selection is largely determined by the experience of a data scientist. However, in the case of an inexperienced data scientist, an algorithm is selected through meta-learning based on dataset characteristics. However, since the selection process is a black box, it was not possible to know on what basis the existing algorithm recommendation was derived. Accordingly, this study uses k-means cluster analysis to classify types according to data set characteristics, and to explore suitable classification algorithms and methods for resolving class imbalance. As a result of this study, four types were derived, and an appropriate class imbalance resolution method and classification algorithm were recommended according to the data set type.

• The Journal of the Korea Contents Association
• /
• v.20 no.8
• /
• pp.653-665
• /
• 2020

As the use of recommended agents has become more active, the "Explain Facilities" is drawing attention as a way to solve the black-box problem that could not explain internal logic to users. This study wants to look at how the description Method and Detail affects to user experience. The Explanation method was divided into 'why the agent did a particular action' and 'why not do a particular action' and the detail condition were divided into 'high or low'. Studies have found that 'why method' have a positive effect on users' transparency, trust, satisfaction, and behavioral intention to use, and 'high detail condition' higher the user' Psychological reactance. In addition, it was found that the explanation methods and detail influenced the 'Explanation' perception through interaction and tended to affect satisfaction and intention to adopt recommendation. This study suggested that careful attention is needed to determine the method and detail of the Explanation facilities in the context of the recommended agent, based on the research findings that it affects the user experience through the interaction of the method and detail.

• Composites Research
• /
• v.31 no.1
• /
• pp.12-16
• /
• 2018

Composite material and mold steel can be expanded differently with the temperature gradients during the forming process because their coefficients of thermal expansions are not the same. Therefore, in order to manufacture the product with accuracy, it is necessary to verify that the forming pressure on the surface of the composite material is maintained to the required level from the material supplier. In this paper, the pressure between the composite material and mold due to the temperature difference was predicted by finite element analysis and the accuracy of predicted value was verified by measuring the thermal expansions of mold steel by the ruler. The predicted value by finite element analysis is closely in agreement with one by the experiment within the required tolerance value of ${\pm}0.05mm$.