KJC Korea
Journal Central

In this paper, we propose a training method of a recommendation learning model that analyzes the battlefield situation and recommends a suitable hypothesis for the current situation. The proposed learning model uses the preference determined by comparing the two hypotheses as a label data to learn which hypothesis best analyzes the current battlefield situation. Our model is based on Siamese neural network architecture which uses the same weights on two different input vectors. The model takes two hypotheses as an input, and learns the priority between two hypotheses while sharing the same weights in the twin network. In addition, a score is given to each hypothesis through the proposed post-processing ranking algorithm, and hypotheses with a high score can be recommended to the commander in charge.

In the IoT environment, various objects mutually interactive, and various services can be composed based on this environment. In the previous study, we have developed a resource collaboration system to provide services by substituting limited resources in the user's personal device using resource collaboration. However, in the preceding system, when the number of resources and situations increases, the inference time increases exponentially. To solve this problem, this study proposes a method of classifying users and resources by applying the BIG5 user type classification model. In this paper, we propose a method to reduce the inference time by filtering the user's preferred resources through BIG5 type-based preprocessing and using the filtered resources as an input to the recommendation system. We implement the proposed method as a prototype system and show the validation of our approach through performance and user satisfaction evaluation.

In this paper, we propose a several smoothing techniques are compared and applied to increase the application of the LSTM-based learning model and its effectiveness. The applied smoothing technique is Savitky-Golay, exponential smoothing, and weighted moving average. Through this study, the LSTM algorithm with the Savitky-Golay filter applied in the preprocessing process showed significant best results in prediction performance than the result value shown when applying the LSTM model to Bitcoin data. To confirm the predictive performance results, the learning loss rate and verification loss rate according to the Savitzky-Golay LSTM model were compared with the case of LSTM used to remove complex factors from Bitcoin price prediction, and experimented with an average value of 20 times to increase its reliability. As a result, values of (3.0556, 0.00005) and (1.4659, 0.00002) could be obtained. As a result, since crypto-currencies such as Bitcoin have more volatility than stocks, noise was removed by applying the Savitzky-Golay in the data preprocessing process, and the data after preprocessing were obtained the most-significant to increase the Bitcoin prediction rate through LSTM neural network learning.

In this paper, we propose a conversational AI agent based on continual learning that can continuously learn and grow with new data over time. A continual learning-based conversational AI agent consists of three main components: Task manager, User attribute extraction, and Auto-growing knowledge graph. When a task manager finds new data during a conversation with a user, it creates a new task with previously learned knowledge. The user attribute extraction model extracts the user’s characteristics from the new task, and the auto-growing knowledge graph continuously learns the new external knowledge. Unlike the existing conversational AI agents that learned based on a limited dataset, our proposed method enables conversations based on continuous user attribute learning and knowledge learning. A conversational AI agent with continual learning technology can respond personally as conversations with users accumulate. And it can respond to new knowledge continuously. This paper validate the possibility of our proposed method through experiments on performance changes in dialogue generation models over time.

In this paper, we propose a method to measure the similarity between golf swings in videos. As it is known that deep learning-based artificial intelligence technology is effective in the field of computer vision, attempts to utilize artificial intelligence in video-based sports data analysis are increasing. In this study, the joint coordinates of a person in a golf swing video were obtained using a deep learning-based pose estimation model, and based on this, the similarity of each swing segment was measured. For the evaluation of the proposed method, driver swing videos from the GolfDB dataset were used. As a result of measuring swing similarity by pairing swing videos of a total of 36 players, 26 players evaluated that their other swing sequence was the most similar, and the average ranking of similarity was confirmed to be about 5th. This ensured that the similarity could be measured in detail even when the motion was performed similarly.

The Global Positioning System (GPS) was developed for military purposes and developed as it is today by opening civilian signals (GPS L1 frequency C/A signals). The current satellite orbits the earth about twice a day to measure the position, and receives more than 3 satellite signals (initially, 4 to calculate even the time error). The three-dimensional position of the ground receiver is determined using the data from the radio wave departure time to the radio wave Time of Arrival(TOA) of the received satellite signal through trilateration. In the case of navigation using GPS in recent years, a location error of 5 to 10 m usually occurs, and quite a lot of areas, such as apartments, indoors, tunnels, factory areas, and mountainous areas, exist as blind spots or neutralized areas outside the error range of GPS. Therefore, in order to acquire one's own location information in an area where GPS satellite signal reception is impossible, another method should be proposed. In this study, IMU(Inertial Measurement Unit) combined with an acceleration and gyro sensor and a geomagnetic sensor were used to design a system to enable location recognition even in terrain where GPS signal reception is impossible. A method to track the current position by calculating the instantaneous velocity value using a 9-DOF IMU and a geomagnetic sensor was studied, and its feasibility was verified through production and experimentation.

In this paper, we propose a function and service of the Disaster Crisis Alert Management System that automatically analyzes the situation judgment criteria to issue a disaster crisis alert and a plan to operate in the National Disaster Management System(NDMS). In the event of a disaster, a crisis alert(interest-caution-alert-serious) is issued according to the crisis alert level. In order to automatically analyze and determine the crisis alert level, first, data collection, crisis alert level analysis, crisis alert level judgment, and disaster crisis alert management system that expresses the crisis alert level by spatial scale(province, city, district) were implemented. The crisis alert level was analyzed and expressed in two ways by applying the intelligent crisis alert level(determination of regional sensitivity, risk level, and crisis alert level) and the crisis alert standard of the crisis management manual(province-level standard setting). Second, standard metadata, linkage of situation information of target) and API standards for data provision are presented to jointly utilize data linkage and crisis alert data of the disaster and safety data sharing platform so that it can be operated within the NDMS.

In this paper, we propose a CCTV storage space securing model using YOLO v3. CCTV is installed and operated in various parts of society for disasters, disasters and safety such as crime prevention, fire prevention, and monitoring, and the number of CCTV is increasing and the quality of the video quality is improving. Due to this, as the number and size of image files increase, it is difficult to cope with the existing storage space. In order to solve this problem, we propose a model that detects specific objects in CCTV images using YOLO v3 library and deletes unnecessary frames by saving only the corresponding frames, thereby securing storage space by reducing the size of the image file, and thereby Periodic images can be stored and managed. After applying the proposed model, it was confirmed that the average image file size was reduced by 94.9%, and it was confirmed that the storage period was increased by about 20 times compared to before the application of the proposed model.

In this paper, we analyze the structure of TV video software, one of the warship combat management system software, and propose a standard architecture that minimizes software modification due to the discontinuation replacement of warship hardware. The class structure was newly designed to minimize the class modified when replacing the warship hardware by separating the variable elements and common elements of TV video software through FORM(Feature-Oriented Reuse Method), the common part that communicates with the warship combat management system and displays the TV screen and the variable part that communicates between the operator and the TV camera. In addition, the Strategy design pattern is applied to efficiently add and modify classes that directly use hardware-dependent APIs when replacing hardware discontinuation, and to make both discontinued and replacements available software. Finally, the reliability testing time and functional testing time of the existing TV video software and the proposed software were measured and compared, and finally, it was confirmed that the hardware discontinuation replacement cost was reduced.

Construction is one of the oldest fields of providing human living comfort. However, despite the continuous development not only in IT but also in other fields, productivity degradation due to low automation is still serious. In particular, the use of the 4th industrial revolution technologies in construction is just beginning. Blockchain, which is one of the 4th industrial revolution technologies, is a type of a distributed database that is used to replicate, share, and synchronize data spread across different geographical locations, such as multiple sites, countries, or organizations. In this paper, we introduce the block chain and analyze use cases according to seven themes in construction. As a result of the analysis, it is expected to increase the understanding of blockchain in construction and provide usefulness for the development of various services in the future.

With the continuous development of the speech recognition system, the recognition rate for speech has developed rapidly, but it has a disadvantage in that it cannot accurately recognize the voice due to the noise generated by mixing various voices with the noise in the use environment. In order to increase the vocabulary recognition rate when processing speech with environmental noise, noise must be removed. Even in the existing HMM, CHMM, GMM, and DNN applied with AI models, unexpected noise occurs or quantization noise is basically added to the digital signal. When this happens, the source signal is altered or corrupted, which lowers the recognition rate. To solve this problem, each voice In order to efficiently extract the features of the speech signal for the frame, the MFCC was improved and processed. To remove the noise from the speech signal, the noise removal method using the Gaussian model applied noise deviation estimation was improved and applied. The performance evaluation of the proposed model was processed using a cross-correlation coefficient to evaluate the accuracy of speech. As a result of evaluating the recognition rate of the proposed method, it was confirmed that the difference in the average value of the correlation coefficient was improved by 0.53 dB.

In this paper, we propose a method of creating a 3D bounding box for an object using a vanishing point to increase the accuracy of object recognition in an image when recognizing an traffic object using a video camera. Recently, when vehicles captured by a traffic video camera is to be detected using artificial intelligence, this 3D bounding box generation algorithm is applied. The vertical vanishing point (VP1) and horizontal vanishing point (VP2) are derived by analyzing the camera installation angle and the direction of the image captured by the camera, and based on this, the moving object in the video subject to analysis is specified. If this algorithm is applied, it is easy to detect object information such as the location, type, and size of the detected object, and when applied to a moving type such as a car, it is tracked to determine the location, coordinates, movement speed, and direction of each object by tracking it. Able to know. As a result of application to actual roads, tracking improved by 10%, in particular, the recognition rate and tracking of shaded areas (extremely small vehicle parts hidden by large cars) improved by 100%, and traffic data analysis accuracy was improved.

The purpose of this study is to improve the operation and working environment of welfare centers through this study on emotional labor of Certified Nursing Assistants at home welfare centers for the elderly. For the study, 138 Certified Nursing Assistants belonging to the Home elderly welfare center in K-si and J-do were studied. For statistical analysis, Cronbach’s α, t-test, ANOVA, Pearson correlation analysis, and multiple regression analysis were performed using SPSS/WIN 17.0. First, as a result of the study, the difference in emotional labor among Certified Nursing Assistants at home-based senior welfare centers was higher in the degree of emotional labor in the order of dementia education, work experience, employment type, education level, and age. Second, the degree of burnout of the survey subjects was highest in the order of work experience, dementia education, employment type, and age. Finally, work stress and emotional labor were found to have a significant positive effect on work burnout. In conclusion, it is necessary to study Certified Nursing Assistants' cases of systematic emotional control experience to prevent work burnout, dementia education, work feedback on how to cope with services, and work conditions improvement measures to relieve work stress.

This study aimed to analyze the changes in news articles related to 'Elderly Living Alone' by comparing Big Data-based news articles related to 'Elderly Living Alone' reported before and after the outbreak of COVID-19. For this, 2018 to 2019 were selected before the outbreak of COVID-19, and 2020 to 2021 were selected after the outbreak, and news articles related to 'Elderly Living Alone' were collected and analyzed using BIGKinds. The main results are as follows. First, the number of related articles decreased after the outbreak of COVID-19 compared to before. Second, there was no significant difference in the analysis of related words. Third, in the relationship diagram analysis, ‘Executives’ before the outbreak of COVID-19 and ‘Corona 19’ after that showed the most weight. This study is expected to be used as basic data in preparing improvement plans for national policies and systems in the context of the spread of infectious diseases in relation to 'Elderly Living Alone'.

In this study, the design conditions and CFD analysis results are compared and reviewed in SCR that can optimally reduce nitrogen compounds. To this end, it was analyzed and compared using CFD to see if the design criteria were satisfied for the shell and tube areas of the boiler. In the SCR system, the analysis area is the gas/air heat exchanger on the shell side, and eight tubes of the gas/air heat exchanger on the tube side. Through CFD analysis, the gas velocity distribution on the primary catalyst side of the SCR system was designed to be 2.4%, and the NH3/NOx molar ratio distribution was 3.7%, which satisfied the design criteria. In addition, the uniformity of the temperature distribution was confirmed and the required condition of 260°C or higher was satisfied. The angle of the gas entering the catalyst met the design conditions at 2.9 degrees, and the pressure loss that occurred also satisfied the design requirements. Through this CFD analysis, it was confirmed that it was designed and operated by satisfying the design conditions required for each area.

In this paper, we propose the positive effects of Virtual Reality(VR) sports classes and the foundation for VR sports to become the basis of lifelong sports education through the application of physical education classes in sports virtual reality programs are to be provided. For this purpose, the effect of VR sports experience on sports satisfaction, sports immersion, and sports attitude factors was investigated for 281 elementary school students in Busan. Results It was found that VR sports experience had a significant effect on sports satisfaction, sports satisfaction had a significant effect on sports immersion and sports attitude, and sports immersion had a significant effect on sports attitude. The great advantage of sports virtual reality is that sports activities for items that are difficult to deal with in physical education classes and unpopular items will be easily performed. In addition, by using a program that links physical education classes with English and mathematics, physical education will be recognized as a convergence subject by elementary school students, and at the same time, it will become an integrated subject that can acquire fun elements and learning elements at the same time through play or games.

In This paper, we propose a determine the correlation between bat swing speed and balance ability in elementary school baseball players. This study subjects were 40 subjects. Bat swing speed was measured using a zepp2baseball, and balance ability was measured using a biorescue. For the measurement of balance ability, eyes open surface·eyes open length·eyes open average speed·eyes close surface·eyes close length·eyes close average speed were measured. pearson's correlation analysis was used to analyze the correlation between bat swing speed and balance ability. There was a significant negative correlation between bat swing speed and eyes close surface (r=-0.366, p<0.05). There was a significant negative correlation between bat swing speed and eyes close length(r=-0.348, p<0.05). There was a significant negative correlation between bat swing speed and eyes close average speed(r=-0.455, p<0.01). Among them, eyes close average speed showed the highest correlation (p<0.01). These findings suggest that improvement of the balance ability has a positive effect on bat swing speed.

This paper aims to investigate the relationship between consumption value of sports products, upward consumption propensity, and post-purchase happiness among adolescents. The subjects of the study were extracted from adolescents in Chungcheong-do by convenience sampling method, and 257 people were used in the final analysis. Statistical methods were frequency analysis, correlation analysis, and regression analysis. The research results are as follows. First, social value and exploratory value, which are sub-factors of consumption value, were shown to have a significant effect on upward consumption propensity, but functional value and self-value were found to be statistically insignificant. Second, social value, self-value, and functional value, which are sub-factors of consumption value, had a significant effect on post-purchase happiness, but exploratory value was found to be statistically insignificant. Third, the upward comparison propensity was found to have a significant effect on post-purchase happiness.

In the digital age, innovation is at the heart of information technology development. This paper explores the differences of regional innovation capabilities from within the innovation system from a new perspective. By applying the coupling coordination degree model, the coupling coordination degree analysis is conducted for the three systems of innovation (innovation input, innovation output, and innovation environment) in 31 provinces, municipalities, and autonomous regions of China. The results show that: the overall level of coupling coordination degree in China is low, and the number of provinces in good coordination and quality coordination is low number, but they are all distributed in the eastern economically developed regions, showing a high correlation between regional innovation and regional economic development. The overall trend of China's innovation coordination is on the rise. The research results of this paper can provide a new perspective for the evaluation of innovation capability, and also provide an important impact of innovation on promoting the development of ICT industry.

In this study, based on existing studies on music streaming services and e-services, the selection factors for music streaming platforms were derived, and the AHP technique was implemented to calculate the importance of each factor. As a result of this study, economic feasibility was found to be the most important factor among security, economic feasibility, informativeness, convenience, and responsiveness, which are the first-step selection factors of music streaming platforms. As a result of synthesizing the weights of the first and second factors, reasonable price was found to be the most important factor. Finally, an additional analysis was conducted to determine whether there was a difference in importance between the selection factors of the music streaming platform according to gender and age. Through this study, it will be possible to figure out the factors that consumers consider most important when using a music streaming platform.

As the risk of privacy invasion due to self-disclosure increases in SNS environment, many studies have tried to discover the influencing factors of self-disclosure. This study is an extension of this research stream and pays attention to the role of interaction privacy controls(friend list and privacy settings) as a new influencing factor. Specifically, the study theorizes and test the logic that the ability to effectively control interactions between individuals using IPC(called IPC usefulness) satisfies the three psychological needs(autonomy, relationship, and competency needs) suggested by the Self-Determination Theory, and in turn increase the amount of self-disclosure. As a result of data analysis, it was found that IPC usefulness has a very strong influence on the satisfaction of psychological needs and is a major factor in increasing the degree of self-disclosure by users. Based on these findings, the study discusses the theoretical and practical implications as well as future research directions.

The purpose of this study was to develop and apply an online-based Arduino programming experience program. In this study, the program was developed and applied based on the contents presented in the ADDIE model and Lee's model. The results of this study were as follows. First, two types of educational program were developed for online use by students of the free semester system. Second, the program content included practical materials such as teaching websites and video lectures. Third, the developed program was applied a total of three times in the school environment and when the student did not attend school. Fourth, analysis of student satisfaction confirmed that the students were satisfied with the development program. The content of this study will serve as a reference for instructors who plan online-based hands-on activities in places where instructors are not readily available, such as islands or mountainous areas.

This study aimed to identify the development direction of Arduino-based boards relating to artificial intelligence based on social awareness identified using big data analytical methods. For the purpose, big data were extracted through the Textom website, focusing on keywords that included 'Arduino + artificial intelligence' and 'Arduino + AI', and these data were refined and analyzed using the Textom website and the UNICET program. In this study, big data analyses, including frequency analysis, TF-IDF analysis, Degree Centrality analysis, N-gram analysis, and CONCOR analysis, were performed. The analyses’ results confirmed that keywords relating to education and coding education, keywords relating to making and experience based on Arduino, and keywords relating to programs were the main keywords used in Arduinoand artificial intelligence-related Internet documents, and clusters were formed based on these keywords confirmed. The social awareness of Arduino and artificial intelligence was evaluated, and the direction of board development was identified based on this social awareness. This study is meaningful in that it identified various factors of board development based on the general public's social awareness, which was evaluated using a big data analysis method. This study may serve as a point of reference for future researchers or developers wishing to understand user needs using big data analysis methods.

As the demand for AI education increases, AI education is actively conducted in the educational field, but it is difficult to internalize AI education due to securing time, difficulty in organizing class contents, and lack of curriculum. As a way to solve this problem, there is a school autonomous course. The school autonomous course allows schools to have autonomy and discretion throughout the curriculum, such as adjusting the number of hours in the subject group and restructuring the use of achievement standards. In this study, in order to enhance AI education, the effect was analyzed by developing and applying an AI education immersion course using a school autonomous curriculum. In the AI education immersion course, students continuously experience AI education in a dense manner within a limited time, so substantial AI education can be achieved. After the AI curriculum, it was found that students' overall AI literacy and self-determination learning motivation improved. It is expected that this study will be able to present a direction to internalize AI education using school autonomous curriculum.