KJC Korea
Journal Central

In this paper, we propose the Deep Learning-Based Companion Animal Abnormal Behavior Detection Service, which using video and sensor data. Due to the recent increase in households with companion animals, the pet tech industry with artificial intelligence is growing in the existing food and medical-oriented companion animal market. In this study, companion animal behavior was classified and abnormal behavior was detected based on a deep learning model using various data for health management of companion animals through artificial intelligence. Video data and sensor data of companion animals are collected using CCTV and the manufactured pet wearable device, and used as input data for the model. Image data was processed by combining the YOLO(You Only Look Once) model and DeepLabCut for extracting joint coordinates to detect companion animal objects for behavior classification. Also, in order to process sensor data, GAT(Graph Attention Network), which can identify the correlation and characteristics of each sensor, was used.

In this paper, a model that can increase the efficiency of work in arranging interior furniture by applying augmented reality technology was studied. In the existing system to which augmented reality is currently applied, there is a problem in that information is limitedly provided depending on the size and nature of the company's product when outputting the image of furniture. To solve this problem, this paper presents an AR labeling algorithm. The AR labeling algorithm extracts feature points from the captured images and builds a database including indoor location information. A method of detecting and learning the location data of furniture in an indoor space was adopted using the CNN technique. Through the learned result, it is confirmed that the error between the indoor location and the location shown by learning can be significantly reduced. In addition, a study was conducted to allow users to easily place desired furniture through augmented reality by receiving detailed information about furniture along with accurate image extraction of furniture. As a result of the study, the accuracy and loss rate of the model were found to be 99% and 0.026, indicating the significance of this study by securing reliability. The results of this study are expected to satisfy consumers' satisfaction and purchase desires by accurately arranging desired furniture indoors through the design and implementation of AR labels.

In this paper, we propose the Detector Scheduler which determines the best tracking-by-detection (TBD) mechanism to perform real-time high-accurate multi-object tracking (MOT). The Detector Scheduler determines whether to run a detector by measuring the dissimilarity of features between different frames. Furthermore, we propose a self-supervision method to learn the Detector Scheduler with tracking results since it is difficult to generate ground truth (GT) for learning the Detector Scheduler. Our proposed self-supervision method generates pseudo labels on whether to run a detector when the dissimilarity of the object cardinality or appearance between frames increases. To this end, we propose the Detector Scheduling Loss to learn the Detector Scheduler. As a result, our proposed method achieves real-time high-accurate multi-object tracking by boosting the overall tracking speed while keeping the tracking accuracy at most.

Recently, with the development of deep learning technology, a variety of huge models with excellent performance have been devised by pre-training massive amounts of text data. However, in order for such a model to be applied to real-life services, the inference speed must be fast and the amount of computation must be low, so the technology for model compression is attracting attention. Knowledge distillation, a representative model compression, is attracting attention as it can be used in a variety of ways as a method of transferring the knowledge already learned by the teacher model to a relatively small-sized student model. However, knowledge distillation has a limitation in that it is difficult to solve problems with low similarity to previously learned data because only knowledge necessary for solving a given problem is learned in a teacher model and knowledge distillation to a student model is performed from the same point of view. Therefore, we propose a heterogeneous knowledge distillation method in which the teacher model learns a higher-level concept rather than the knowledge required for the task that the student model needs to solve, and the teacher model distills this knowledge to the student model. In addition, through classification experiments on about 18,000 documents, we confirmed that the heterogeneous knowledge distillation method showed superior performance in all aspects of learning efficiency and accuracy compared to the traditional knowledge distillation.

In this paper, we propose a method of analyzing research papers published by researchers belonging to university departments in the field of disaster & safety for the scientometric analysis of the research status in the field of disaster safety. In order to conduct analysis research, the dataset constructed in previous studies was newly improved and utilized. In detail, for research papers of authors belonging to the disaster prevention and safety engineering type department of domestic universities, institution identification, cited journal identification of references, department type classification, disaster safety type classification, researcher major information, KSIC(Korean Standard Industrial Classification) mapping information was reflected in the experimental data. The proposed method has a difference from previous studies in the field of disaster & safety and data set based on related keyword searches. As a result of the analysis, the type and regional distribution of organizations belonging to the department of disaster prevention and safety engineering, the composition of co-authored department types, the researchers' majors, the status of disaster safety types and standard industry classification, the status of citations in academic journals, and major keywords were identified in detail. In addition, various co-occurrence networks were created and visualized for each analysis unit to identify key connections. The research results will be used to identify and recommend major organizations and information by disaster type for the establishment of an intelligent crisis warning system. In order to provide comprehensive and constant analysis information in the future, it is necessary to expand the analysis scope and automate the identification and classification process for data set construction.

This research conducted a study classifying gender of speakers by analyzing feature vectors extracted from the voice data. The study provides convenience in automatically recognizing gender of customers without manual classification process when they request any service via voice such as phone call. Furthermore, it is significant that this study can analyze frequently requested services for each gender after gender classification using a learning model and offer customized recommendation services according to the analysis. Based on the voice data of males and females excluding blank spaces, the study extracts feature vectors from each data using MFCC(Mel Frequency Cepstral Coefficient) and utilizes SVM(Support Vector Machine) models to conduct machine learning. As a result of gender classification of voice data using a learning model, the gender recognition rate was 94%.

In this study, I analyzed the inducing and hindering factors of online shopping malls on browsers in the process of browsing fashion products in online shopping malls. In-depth interviews were conducted by presenting browsing-related questions to women in their 20s and 50s who are interested in browsing fashion products online. Based on the answers of the interviewees, using grounded theory, we analyzed and presented six factors such as price factor, promotion factor, purchase review factor, visual information factor, product information factor, and service factor. Based on inducing and hindering factors to browsing analyzed in this study, a strategy to design a browsing environment in terms of shopping malls was suggested, which will be helpful for practical strategies and marketing in related industries. Basic data will be presented in a thesis on a new type of shopping mall browsing environment related to the rapidly developing information and communication technology. In addition, the negative emotions experienced in relation to the detrimental factors of shopping malls in the browsing process are expected to be helpful in researching fashion product browsing related to consumer psychology.

The purpose of this study is to present supporting data that can suggest a direction for sports advertisement strategy on social media by investigating how differences in social media engagement of sports advertisement affect users' product trust and purchase intention. To achieve this purpose, a total of 810 questionnaires, which were completed by sports-related college students engaged in social media activities using a method for writing a self-evaluation report, were collected, and 765 questionnaires were used as final valid samples after excluding 45 insincerely completed copies. For data processing, analyses of frequency, factor, reliability, correlation, and multiple regression were conducted using the Window SPSS (ver. 23.0) statistical program to draw the following conclusions. First, the social media engagement of the sports advertisement had a significant effect on the trust of the advertised products. Second, the social media engagement of the sports advertisement had a significant effect on the purchase intention of the advertised products. Third, the product trust acquired through the sports advertisement had a significant effect on the purchase intention.

This paper proposes a computationally efficient 2-D direction-of-arrival (DoA) estimation method with a uniform rectangular array (URA). This method is called the direct signal-based method in the sense that it is based directly on the phase relationships among the signals arriving at each antenna of an antenna array rather than their correlation matrix. According to the simulation results, it has be shown that the direct signal-based method, with much less computations than any existing methods, yields the performance comparable to that of the MUSIC (MUltiple SIgnal Classification) method in terms of the root-mean-squared error (RMSE) and the maximum absolute error.

As 5G mobile communication services gradually expand in P2P (peer-to-peer) or D2D (device-to-device) applications, traffic-oriented stream control such as YouTube streaming is emerging as an important technology. In D2D communication, the type of data stream most frequently transmitted by users is a video stream, which has the characteristics of a large-capacity transport stream. In a D2D communication environment, this type of stream not only provides a cause of traffic congestion, but also degrades the quality of service between D2D User Equipments (DUEs). In this paper, we propose a Traffic-Oriented Stream Scheduling (TOSS) scheme to minimize the interruption of dynamic media streams such as video streams and to optimize streaming service quality. The proposed scheme schedules the media stream by analyzing the characteristics of the media stream and the traffic type in the bandwidth of 3.5 GHz and 28 GHz under the 5G gNB environment. We examine the performance of the proposed scheme through simulation, and the simulation results show that the proposed scheme has better performance than other comparative methods.

In this paper, we present and analyze 5G system and IEEE time-sensitive networking(TSN) and propose a novel mobility management scheme for time sensitive communications in 5G-TSN to support ultra-low latency networks. Time-sensitive networking(TSN) has a promising future in the Industrial Automation and Industrial Internet of Things(IIoT), as a key technology that is able to provide low-latency, high-reliable and deterministic communications over the Ethernet. When a TSN capable UE moves the TSN service coverage from the non-TSN service coverage, the UE cannot get the TSN service promptly because the related mobility management is not performed appropriately. For the mobility situation with the TSN service coverage, the proposed scheme reports TSN capability to the network and triggers the initial registration in order to be provided the TSN service immediately and ultra-low latency communications compared to existing schemes in 5G mobile networks.

In this paper, we analyze trends in the use of mobile forensic technology, focusing on cases where mobile forensics are used, and we predict the development of future mobile forensics technology using linear regression used in future prediction models. For the current status and outlook analysis, we extracted a total of 8 variables by analyzing 1,397 domestic and foreign mobile forensics-related cases and newspaper articles. We analyzed the prospects for each variable using the year of occurrence as an independent variable, seven variables such as text (text message usage information), communication information (cell phone communication information), Internet usage information, messenger usage information, stored files, GPS, and others as dependent variables. As a result of the analysis, among various aspects of the use of mobile devices, the use of Internet usage information, messenger usage information, and data stored in mobile devices is expected to increase. Therefore, it is expected that continuous research on technologies that can effectively extract and analyze characteristic information of mobile devices such as file systems, the Internet, and messengers will be needed As mobile devices increase performance and utilization in the future and security technology.

In this paper is one of the measures to prevent concrete collapse accidents at construction sites in advance. Analyzed based on accumulated Meteorological Agency data. It is a reliable model that confirms the prediction of the decrease rate occurrence interval, and the verification items such as p_value is 0.5 or less and ecof appears in one direction through the SARIMA model, which is suitable for regular and clear time series data models, ensure reliability. Significant results were obtained. As a result of analyzing the temperature change by time zone and the water reduce ratio by section using the data secured based on such trust, the water reduce ratio is the highest in the 29-31 ℃ section from 12:00 to 13:00 from July to August. found to show. If a factor in the research result interval occurs using the research results, it is expected that the batch plant will produce Ready-mixed concrete that reflects the water reduce ratio at the time of designing the water-cement mixture, and prevent the decrease in concrete compressive strength due to the water reduce ratio.

There are many difficulties and inconveniences in the process of completing driving education for the disabled. Even in the content of driving education, various unpredictable unexpected situations occur during actual driving. Although education on emergency situations is being conducted in the current driving education course, education to cope with various unexpected situations is insufficient. Therefore, in this paper, we propose a Unity-based driving education simulator for the disabled. Create a simulator by composing scenarios for various unexpected situations using Unity. A simulation is conducted through the test subjects, and the coping score for the existing unexpected situation is evaluated according to the evaluation items. It was confirmed that the evaluation score of the ability to cope with unexpected situations of the proposed system was higher than that of the existing driving education simulator. If the proposed system is used, it can be expected to improve the ability to cope with unexpected situations and improve the quality of life through realistic driving experience.

In relation to the travel itinerary recommendation service, which has recently become in high demand, our previous work introduces a method to quantify the popularity of places including tour spots, restaurants, and accommodations through social big data analysis, and to create a travel schedule based on the analysis results. On the other hand, the generated schedule was mainly composed of travel routes that connected tour spots with the shorted distance, and detailed schedule information including restaurants and accommodation information for each travel date was not provided. This paper presents an algorithm for constructing a detailed travel route using a scenario template in a travel schedule created based on social big data, and introduces a prototype system that implements it. The proposed system consists of modules such as place information collection, place-specific popularity score estimation, shortest travel rout generation, daily schedule organization, and UI visualization. Experiments conducted based on social reviews collected from 63,000 places in the Gyeongnam province proved effectiveness of the proposed system.

Most of the metaverse service is a gamification type of metaverse solution composed by placing 3D objects on a 360 panoramic world base. However, the metaverse webtoon service is lacking in implementation and research. In this study, a method for realizing a 2D flat image form in a 3D space and a realistic 360-degree panoramic webtoon metaverse service were studied. The research process consisted of basic storytelling and design work for webtoon production, panoramic image creation to convert the produced image into 360-degree form, and content creation process for viewing 360-degree directions. Finally, we implemented shading and material work with game engine tools so that you can enjoy virtual reality-based webtoons. The webtoon-based metaverse study studied the process of creating a 360-degree panoramic webtoon from content that can only be viewed in 2D format. Accordingly, it is thought that webtoons will also be presented as a standard for the production and implementation method for the metaverse service.

In this paper, we examined the research trends and characteristics related to the metaverse in global journals published between 2000 and 2022 from the Web of Science database. The analysis included descriptive statistics, multidimensional scaling, keyword network analysis, and visualization. In addition, semantic network models were constructed, and centrality (betweenness and degree) analysis was performed using R and KH coder in two separate categories based on the trends and aspects of the publication: analysis period 1 (Jan 2000 to Dec 2020) and period 2 (Jan 2021 to Jun 2022). The results showed that the recent global research trends related to the metaverse could be quantitatively characterized using the semantic network analysis. Also, the results could be applied to suggest future research topics in the field of metaverse based on quantitative and empirical data.

This study attempted to categorize the lower body torso type and investigated its characteristics for women in their 40s, 50s, and 60s, who increase the frequency of incontinence. This study analyzed the 8th human dimension survey data of Korean Agency for Technology and Standards. The data were analyzed by SPSS 26.0 program. It was analyzed that the height item of the lower body torso decreased, the width item widen as the age increases, and the waist and abdomen circumference of the circumference item increased and the hip circumference decreased. Body length decreased with age. The components of the lower torso were classified into the lower torso horizontal factor, height factor, lower factor, and vertical factor. The lower body torso type was classified into a long inverted triangular body type, a short and high body type, a body fat body type, and a low triangular body type. It was analyzed that type 3 was the largest in the horizontal factor and height factor, and type 1 was the largest in the vertical factor and the lower part factor. A new drafting method was required in setting the horizontal part of the incontinence panty, the front and the back length.

This study was conducted to analyze the effect of dance exercise in Aqua(AZ) and Land(ZG) environments on bone metabolism marker(BMM) and bone mineral density(BMD) and to suggested an effective exercise method to prevent osteoporosis and osteopenia. The subjects of the study were randomly assigned to an aqua zumba group (AZG, n=11), a zumba gold group (ZGG, n=13), and a control group (CG, n=12) for elderly women over 70 years age and 12 weeks, dance exercises were conducted twice time a week for 60 minutes. As a result of this study, OC was AZG(P<.001) and ZGG(P<.05), and IGF-1 was significantly increased in AZG(P<.01). However, there was no significant difference in DPD, BMD, and T-score. Therefore, dance (Zumba) exercise in Aqua and Land has positive effects on BMM and BMD in elderly women, so it is suggested as an effective intervention method to delay osteoporosis and osteopenia.

This study attempted to compare and analyze the commercially available multi-use patterns to develop mask patterns suitable for the face types of adult women. Through this, it was intended to provide necessary data to mask pattern development and products. As a results of comparing the dimensions and shapes of commercial multi-use mask patterns, there was a significant difference in dimensions even though it was a L-size mask manufactured for adults. As a result of the appearance evaluation of the virtual outfit, there were significant differences by design in the vertical of the center front line, the cover and space of the mask, the height of the nose, and the lower part of the mask. The side also showed significant differences in the covering of the side of the face, the space of the side, and the width and length of the string. As a result of the appearance evaluation, Mask 4 received the best evaluation. The shape of the mask pattern had a large dart in the lower part of the nose so that it can cover the three-dimensional shape of the face, but there was a difference in the degree and angle of the curve depending on the mask. Although the upper part of the mask, the lower part of the mask, and the cheek part are in close contact, the evaluation of the mask pattern, which has room in the nose and mouth, was high. It is thought that the mask pattern should be set according to the upper length, lower length, and nose height of the mask through analysis of the face shape and dimensions.

This study explored the effects of Tai Chi on cardiovascular disease(CVD) risk factors among the Korean elderly. 68 elderly were selected and divided into two groups: an experiment group(34 patients) and a control group(34 patients). The experimental group practiced Tai Chi for 60 minutes a day, five times per week for 6 weeks total. The control group continued their routine levels of activity. CVD risk factors evaluated in both groups including body mass index, fasting blood sugar, triglyceride, high density lipoprotein cholesterol(HDL-C), and low density lipoprotein cholesterol(LDL-C). Each factor was measured before and after the 6-week intervention. Results showed that the experimental group had a significant decrease in measurements of body mass index, fasting blood sugar, triglyceride, and LDL-C levels. The experimental group also had an increase HDL-C level. Based on such findings, we believe that Tai Chi exercise program would function as an effective intervention strategy for preventing CVD in the elderly. Therefore, it is recommended that health and medical service personnel should seek to apply Tai Chi intervention as a daily component of elderly care.

The purpose of this study examines to what extent the multicultural youth's dual culture acceptance attitude is significantly affected by cultural adaptation stress and national identity using the data of the MAPS(Multicultural Adolescents Panel Study) conducted by the National Youth Policy Institute. The subject of the study was the first data of the second period of the MAPS, and 2,246 multicultural youth who were enrolled in the fourth grade of elementary school as of 2019 were used as analysis data. As a result of the study, it was found that the attitude to accept dual culture was significantly affected in the order of national identity and cultural adaptation stress. This means that the higher the national identity and the lower the cultural adaptation stress, the higher the attitude toward dual culture. On the other hand, as for the type of multicultural youth, it was found that international marriage families had the lowest attitude toward accepting dual culture. In terms of the size of the area where students live, large cities have the lowest dual cultural acceptance attitude. These results suggest that cultural adaptation stress, national identity, type of multicultural family, and area of residence act as major variables in multicultural youth's dual culture acceptance attitude.

The purpose of this study is to indicate the direction of the future university classes in the post-COVID era, comparing and analyzing lecture evaluation of pre and post COVID-19. To this end, 4 yeard data were used from 2018 to 2019 for pre COVID-19 and form 2020 to 2021 data for post COVID-19. The results were as follows. In the case of liberal arts, “assignments ” was the word with the highest frequency and degree centrality(DC) regardless of pre and post-COVID-19 In the major, “understanding” appeared as the most important word. The result of the ego network analysis indicated that “video lecture” and “non-face-to-face classes” were difficult and “interaction” between the professor and the students was important. As a results, it is important to reduce the weight of assignments and increase interaction with students in liberal arts classes. In the case of majors, it is necessary to operate face-to-face classes rather than non-face-to-face classes, and to organize the contents of videos without difficulty.

In this paper, we examined the causes of information privacy concerns, trust and related factors in social network services. On the basis of the 'elaboration likelihood model,' we established factors such as information quality, privacy policy, perceived SNS app popularity and optimism bias affecting information privacy concern of SNS users. In addition, we analyzed the relationship between information privacy concern, trust in SNS members, trust in SNS platform and intention to use. Finally, on the basis of the 'trust transfer theory', we analyzed the relationship between trust in SNS members and trust in SNS platform. The results of the study showed that (1) information quality, privacy policy and optimism bias had the significant effects on information privacy concerns, (2) perceived SNS app popularity didn’t affect information privacy concerns, (3) information privacy concerns had the significant effects on trust in SNS platforms (4) in accordance with the trust transfer theory, trust in SNS members had the significant effect on trust in SNS platforms, and (5) trust in SNS members had the significant effect on intention to use, while trust in SNS platform didn't have the significant effect on intention to use. The findings of the study are expected to help to improve the SNS firms' understanding towards customers’ information privacy protection behaviors and trust.

This paper aims to find promising export items for market expansion of defense export items. Germany, the UK, and France were selected as export target countries to obtain unstructured forecast data on weapons system acquisition plans for the next ten years by each country. Using the TF-IDF in text mining analysis, keywords that appeared frequently in data from three countries were derived. As a result of this paper, keywords for each country's major acquisition projects drawing. However, most of the derived keywords were related to mainstay weapon systems produced by domestic defense companies in each country. To discover promising export items from text mining, we proposed that the drawn keywords are distinguished as similar weapon systems. In addition, we assort the weapon systems that the three countries will get a plan to acquire commonly. As a result of this paper, it can be seen that the current promising export item is a weapon system related to the information system. Prioritizing overseas demands using key words can set clear market entry goals. In the case of domestic companies based on needs, it is possible to establish a specific entry strategy. Relevant organizations also can provide customized marketing support.

In this study, we derived selection factors for OTT services based on previous studies and presented an hierarchy process analysis model to calculate the weights for each evaluation. To evaluate the OTT selection factors, OTT customer data was used for the analysis. The results showed that of the first stage factors of economic feasibility, system excellence, convenience, diversity, playfulness, and responsiveness, diversity is the most important. When the weights for first stage and second stage factors were synthesized, entertainment was the most important factor. Lastly, this study analyzed whether there is a difference in the weight of the selection factor for the first stage of OTT service according to gender and age. This study shows which factors consumers prioritize when using OTT services.

We compared and analyzed the factors affecting the lecture evaluation of IT subjects, which are mandatory for all students of H University. The purpose of this study is to determine whether lecture satisfaction has a significant correlation with academic achievement, attendance rate, and categories of courses. In this study, we check whether the lecture satisfaction of IT liberal arts subjects that require a lot of computer-based practice differs from that of other liberal arts subjects. We used the 2,149 evaluation data of 12 lectures submitted by 2,322 students in the first and second semesters of year 2019 at University H. As for the lecture evaluation results, in addition to the evaluation scores of the multiple choice questions, the subjective questions were also quantified by classifying the statements submitted by the students into positive and negative types to make the results of the lecture evaluation objective. Our research results show that student group who have the higher attendance rates and academic achievements have higher level of lecture satisfaction and they also use more positive words than negative words in subjective evaluation questions. Students with the lower score use the more negative words, but the ratio between positive and negative words does not differ between groups. Higher attendance rates groups in the basic programming courses and software applications courses have higher lecture satisfaction ratio. But in the intermediate programming courses, the higher attendances rate and the lecture satisfaction do not have any significant relationship. Also students in the intermediate programming courses use more negative words than those in the basic programming courses.

The purpose of this study is to investigate the learning satisfaction and perception of the students who have taken the online classes of the Digital Innovation Sharing University in the field of new energy industry, and to analyze the relationship between the learning satisfaction and core competencies. In order to achieve this purpose, the results of the learning satisfaction survey of a total of 470 students who took the digital innovation sharing university classes in the new energy industry were analyzed in both quantitative and qualitative methods. As a result of quantitative analysis, students' learning satisfaction was generally above average, and they were particularly satisfied with the setting and operation of educational goals. In addition, it was found that students' learning satisfaction had a statistically significant effect on students' core competencies. As a result of the qualitative analysis, it was possible to confirm the direction of improvement of the curriculum in the areas of function, operation, and content. Based on the results of this study, it is expected that the digital innovation sharing university curriculum quality management feedback system will be prepared in the future, and it will be the foundation for the continuous establishment of the curriculum quality management system.