• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.615-616
• /
• 2024

현대 야구에서 타자의 스윙 패턴 분석은 상대 투수가 투구 전략을 수립하는데 상당히 중요하다. 이미지 기반의 인간 포즈 추정(HPE)은 대규모 스윙 패턴 분석을 자동화할 수 있다. 그러나 기존의 HPE 방법은 빠르고 가려진 신체 움직임으로 인해 복잡한 스윙 모션을 정확하게 추정하는 데 어려움이 있다. 이러한 문제를 극복하기 위해 스윙 모션에 대한 사전 정보를 활용하여 야구 선수의 포즈를 보정하는 방법(BPPC)을 제안한다. BPPC는 동작 인식, 오프셋 학습, 3D에서 2D 프로젝션 및 동작 인지 손실 함수를 통해 스윙 모션에 대한 사전 정보를 반영하여 기성 HPE 모델 결과를 보정한다. 실험에 따르면 BPPC는 벤치마크 데이터셋에서 기성 HPE 모델의 2D 키포인트 정확도를 정량적 및 정성적으로 향상시키고, 특히 신뢰도 점수가 낮고 부정확한 키포인트를 크게 보정했다.

• Communications for Statistical Applications and Methods
• /
• v.15 no.4
• /
• pp.563-585
• /
• 2008

This paper presents how to improve the efficiency and accuracy in the pricing and sensitivity evaluation for derivatives, since the need for the evaluation of complicated derivatives is increased. The Monte Carlo(MC) simulation using the quasi random number instead of pseudo random number can improve the elapsed time and accuracy for the valuation of European-type derivatives. However, the quasi MC simulation method has its limit for applying it in the multi-dimensional case such as American-type and path-dependent options due to the increased correlation between dimensions as the dimension of random numbers is increased. In order to complement this problem, we develop a modified method in which correlation values are controlled to be below a pre-specified value. Thus, this method is applicable for the pricing of either derivatives ill which underlying assets or risk factors are several or derivatives having path-dependent or early redemption property. Furthermore, we illustrate that it is important to take an appropriate grid interval for the use of finite difference method(FDM) by applying the FDM to one example of non-symmetrical butterfly spreads.

• Journal of Digital Convergence
• /
• v.19 no.12
• /
• pp.347-352
• /
• 2021

In this paper, we propose a method for finding feature subsets that are effective for classification in an input dataset by using a multi-agent reinforcement learning method. In the field of machine learning, it is crucial to find features suitable for classification. A dataset may have numerous features; while some features may be effective for classification or prediction, others may have little or rather negative effects on results. In machine learning problems, feature selection for increasing classification or prediction accuracy is a critical problem. To solve this problem, we proposed a feature selection method based on reinforced learning. Each feature has one agent, which determines whether the feature is selected. After obtaining corresponding rewards for each feature that is selected, but not by the agents, the Q-value of each agent is updated by comparing the rewards. The reward comparison of the two subsets helps agents determine whether their actions were right. These processes are performed as many times as the number of episodes, and finally, features are selected. As a result of applying this method to the Wisconsin Breast Cancer, Spambase, Musk, and Colon Cancer datasets, accuracy improvements of 0.0385, 0.0904, 0.1252 and 0.2055 were shown, respectively, and finally, classification accuracies of 0.9789, 0.9311, 0.9691 and 0.9474 were achieved, respectively. It was proved that our proposed method could properly select features that were effective for classification and increase classification accuracy.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.49 no.2
• /
• pp.217-227
• /
• 2022

This study aimed to investigate parents' and dentists' perceptions of the national oral health screening for infants and children (OHS), and evaluate the dentists' accuracy of analyzing the results of OHS. A questionnaire survey was conducted, face-to-face or online, for 90 parents of children who received OHS in the Department of Pediatric Dentistry at Yonsei University Dental Hospital and 100 dentists working at local clinic or university dental hospital from May to October 2021. Most parents and dentists were aware of the importance of OHS, and approximately 96.7% of parents were satisfied with OHS. The requirements of parents and dentists about OHS were different. The reasons for having difficulties in explaining parents after OHS and the opinions on appropriate period and number of OHS were disagreed between pediatric dentists and general dentists. Regardless of dentists' major, work experience, elapsed period after taking online education program, the accuracy of the examination result was low. In this study, various opinions of parents and dentists on OHS were collected. Efforts should be made to enhance the oral health of infants and children by considering the requirements of parents and dentists and improving the accuracy of examination results.

• Science of Emotion and Sensibility
• /
• v.20 no.3
• /
• pp.119-130
• /
• 2017

The purpose of the study is discovering the effects of virtual competitors on user in AR (Augment Reality) home training system. Specifically, the current research examined their effects on immersion, perceived competition, and leaning motivation. The paper tested three unexplored relationship. First, introducing virtual competitors in home training system will enhance user's immersion. Second, presenting virtual competitors in home training system will increase user's perceived competition. Third, virtual competitors in home training system will raise user's learning motivation. For empirical analysis, we developed home training system, which could check and give feedback automatically, based on user's posture. Using this AR home training system, the study empirically shows how and why virtual competitors affect users. The results give implications not only on service design; but also on the idea that virtual other could affect user's behavior.

• Journal of Internet Computing and Services
• /
• v.21 no.3
• /
• pp.133-144
• /
• 2020

The number of elderly people with dementia is increasing as fast as the proportion of older people due to aging, which creates a social and economic burden. In particular, dementia care costs, including indirect costs such as increased care costs due to lost caregiver hours and caregivers, have grown exponentially over the years. In order to reduce these costs, it is urgent to introduce a management system to care for dementia patients. Therefore, this study proposes a sensor-based abnormal behavior detection system to manage dementia patients who live alone or in an environment where they cannot always take care of dementia patients. Existing studies were merely evaluating behavior or evaluating normal behavior, and there were studies that perceived behavior by processing images, not data from sensors. In this study, we recognized the limitation of real data collection and used both the auto-encoder, the unsupervised learning model, and the LSTM, the supervised learning model. Autoencoder, an unsupervised learning model, trained normal behavioral data to learn patterns for normal behavior, and LSTM further refined classification by learning behaviors that could be perceived by sensors. The test results show that each model has about 96% and 98% accuracy and is designed to pass the LSTM model when the autoencoder outlier has more than 3%. The system is expected to effectively manage the elderly and dementia patients who live alone and reduce the cost of caring.

• Korean Journal of Cognitive Science
• /
• v.24 no.3
• /
• pp.271-300
• /
• 2013

Human and animals are born with an intuitive ability to determine approximate numerosity. This ability is termed approximate number sense (hereafter, number sense). Evolutionarily, number sense is thought to be an essential ability for hunting, gathering and survival. According to previous research, children with mathematical learning disability have impaired number sense. On the other hand, individuals with more accurate number sense have higher mathematical achievement. These results support the hypothesis that number sense provides a basis for the development of mathematical cognition. Recently, researchers have been examining whether number sense training can lead to enhancement in mathematical achievement and changes in brain activity in relation to mathematical problem solving. Numerosity which basically represents discontinuous quantity is expected to be closely related to continuous quantity such as representations of space and time. A theory of magnitude (ATOM) states that processing of number, space and time is based on a common magnitude system in the posterior parietal cortex, especially the intraparietal sulcus. The present paper introduces current literature and future directions for the study of the common magnitude system.

• The Journal of Bigdata
• /
• v.4 no.2
• /
• pp.23-34
• /
• 2019

Recently, the crime that utilizes the digital platform is continuously increasing. About 140,000 cases occurred in 2015 and about 150,000 cases occurred in 2016. Therefore, it is considered that there is a limit handling those online crimes by old-fashioned investigation techniques. Investigators' manual online search and cognitive investigation methods those are broadly used today are not enough to proactively cope with rapid changing civil crimes. In addition, the characteristics of the content that is posted to unspecified users of social media makes investigations more difficult. This study suggests the site-based collection and the Open API among the content web collection methods considering the characteristics of the online media where the infringement crimes occur. Since illegal content is published and deleted quickly, and new words and alterations are generated quickly and variously, it is difficult to recognize them quickly by dictionary-based morphological analysis registered manually. In order to solve this problem, we propose a tokenizing method in the existing dictionary-based morphological analysis through WPM (Word Piece Model), which is a data preprocessing method for quick recognizing and responding to illegal contents posting online infringement crimes. In the analysis of data, the optimal precision is verified through the Vote-based ensemble method by utilizing a classification learning model based on supervised learning for the investigation of illegal contents. This study utilizes a sorting algorithm model centering on illegal multilevel business cases to proactively recognize crimes invading the public economy, and presents an empirical study to effectively deal with social data collection and content investigation.

• Korean Journal of Cognitive Science
• /
• v.27 no.3
• /
• pp.469-499
• /
• 2016

The present study investigated the cognitive mechanism underlying online processing of Korean syntactic (for example, A bed/a clock belongs to/doesn't belong to the furniture "침대는/시계는 가구에 속한다/속하지 않는다") and lexical negation (for example, A tiger/a butterfly has/doesn't have a tail "호랑이는/나비는 꼬리가 있다/없다") using an ERP(Event-related potentials) technique and a truth-value verification task. 23 Korean native speakers were employed for the whole experiment and 15's brain responses (out of 23) were recorded for the ERP analysis. The behavioral results (i.e. verification task scores) show that there is universal pattern of the accuracy and response time for verification process: True-Affirmative (high accuracy and short latency) > False-Affirmative > False-Negated > True-Negated. However, the components (early N400 & P600) reflecting the immediate processing of a negation operator were observed only in lexical negation. Moreover, the ERP patterns reflecting an effect of truth value were not identical: N400 effect was observed in the true condition compared to the false condition in the lexically negated sentences, whereas Positivity effect (like early P600) was observed in the false condition compared to the true condition in the syntactically negated sentences. In conclusion, the form and location of negation operator varied by languages and negation types influences the strategy and pattern of online negation processing, however, the final representation resulting from different computational processing of negation appears to be language universal and is not directly affected by negation types.