• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.337-357
• /
• 2021

Because the growth of paddy rice is affected by meteorological factors, the selection of appropriate meteorological variables is essential to build a rice yield prediction model. This paper examines the suitability of multiple meteorological datasets for the rice yield modeling in South Korea, 1996-2019, and a hindcast experiment for rice yield using a machine learning method by considering the nonlinear relationships between meteorological variables and the rice yield. In addition to the ASOS in-situ observations, we used CRU-JRA ver. 2.1 and ERA5 reanalysis. From the multiple meteorological datasets, we extracted the four common variables (air temperature, relative humidity, solar radiation, and precipitation) and analyzed the characteristics of each data and the associations with rice yields. CRU-JRA ver. 2.1 showed an overall agreement with the other datasets. While relative humidity had a rare relationship with rice yields, solar radiation showed a somewhat high correlation with rice yields. Using the air temperature, solar radiation, and precipitation of July, August, and September, we built a random forest model for the hindcast experiments of rice yields. The model with CRU-JRA ver. 2.1 showed the best performance with a correlation coefficient of 0.772. The solar radiation in the prediction model had the most significant importance among the variables, which is in accordance with the generic agricultural knowledge. This paper has an implication for selecting from multiple meteorological datasets for rice yield modeling.

• Journal of Korea Water Resources Association
• /
• v.53 no.12
• /
• pp.1159-1172
• /
• 2020

This study applied deep convolution neural network based on U-Net and SegNet using long period weather radar data to very short-term rainfall prediction. And the results were compared and evaluated with the translation model. For training and validation of deep neural network, Mt. Gwanak and Mt. Gwangdeoksan radar data were collected from 2010 to 2016 and converted to a gray-scale image file in an HDF5 format with a 1km spatial resolution. The deep neural network model was trained to predict precipitation after 10 minutes by using the four consecutive radar image data, and the recursive method of repeating forecasts was applied to carry out lead time 60 minutes with the pretrained deep neural network model. To evaluate the performance of deep neural network prediction model, 24 rain cases in 2017 were forecast for rainfall up to 60 minutes in advance. As a result of evaluating the predicted performance by calculating the mean absolute error (MAE) and critical success index (CSI) at the threshold of 0.1, 1, and 5 mm/hr, the deep neural network model showed better performance in the case of rainfall threshold of 0.1, 1 mm/hr in terms of MAE, and showed better performance than the translation model for lead time 50 minutes in terms of CSI. In particular, although the deep neural network prediction model performed generally better than the translation model for weak rainfall of 5 mm/hr or less, the deep neural network prediction model had limitations in predicting distinct precipitation characteristics of high intensity as a result of the evaluation of threshold of 5 mm/hr. The longer lead time, the spatial smoothness increase with lead time thereby reducing the accuracy of rainfall prediction The translation model turned out to be superior in predicting the exceedance of higher intensity thresholds (> 5 mm/hr) because it preserves distinct precipitation characteristics, but the rainfall position tends to shift incorrectly. This study are expected to be helpful for the improvement of radar rainfall prediction model using deep neural networks in the future. In addition, the massive weather radar data established in this study will be provided through open repositories for future use in subsequent studies.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.191-206
• /
• 2022

Recently, it is a de-facto approach to utilize a pre-trained language model(PLM) to achieve the state-of-the-art performance for various natural language tasks(called downstream tasks) such as sentiment analysis and question answering. However, similar to any other machine learning method, PLM tends to depend on the data distribution seen during the training phase and shows worse performance on the unseen (Out-of-Distribution) domain. Due to the aforementioned reason, there have been many efforts to develop domain-specified PLM for various fields such as medical and legal industries. In this paper, we discuss the training of a finance domain-specified PLM for the Korean language and its applications. Our finance domain-specified PLM, KB-BERT, is trained on a carefully curated financial corpus that includes domain-specific documents such as financial reports. We provide extensive performance evaluation results on three natural language tasks, topic classification, sentiment analysis, and question answering. Compared to the state-of-the-art Korean PLM models such as KoELECTRA and KLUE-RoBERTa, KB-BERT shows comparable performance on general datasets based on common corpora like Wikipedia and news articles. Moreover, KB-BERT outperforms compared models on finance domain datasets that require finance-specific knowledge to solve given problems.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.259-278
• /
• 2022

With the spread of Artificial Intelligence (AI), various AI-based services are expanding in the financial sector such as service recommendation, automated customer response, fraud detection system(FDS), credit scoring services, etc. At the same time, problems related to reliability and unexpected social controversy are also occurring due to the nature of data-based machine learning. The need Based on this background, this study aimed to contribute to improving trust in AI-based financial services by proposing a checklist to secure fairness in AI-based credit scoring services which directly affects consumers' financial life. Among the key elements of trustworthy AI like transparency, safety, accountability, and fairness, fairness was selected as the subject of the study so that everyone could enjoy the benefits of automated algorithms from the perspective of inclusive finance without social discrimination. We divided the entire fairness related operation process into three areas like data, algorithms, and user areas through literature research. For each area, we constructed four detailed considerations for evaluation resulting in 12 checklists. The relative importance and priority of the categories were evaluated through the analytic hierarchy process (AHP). We use three different groups: financial field workers, artificial intelligence field workers, and general users which represent entire financial stakeholders. According to the importance of each stakeholder, three groups were classified and analyzed, and from a practical perspective, specific checks such as feasibility verification for using learning data and non-financial information and monitoring new inflow data were identified. Moreover, financial consumers in general were found to be highly considerate of the accuracy of result analysis and bias checks. We expect this result could contribute to the design and operation of fair AI-based financial services.

• Journal of the Korean Dietetic Association
• /
• v.11 no.2
• /
• pp.205-215
• /
• 2005

Health is easily overlooked because it doesn’t be changed good or bad due to sudden effort or indifference unexpectedly but kept in daily life. Especially, schoolchildren period, an important lifetime to develop both physically and mentally needs to be helpful to promote the growth of the body and keep well-balanced mind through balanced and nourishing diet. The purpose of this study was to develop nutrition education contents for discretional activities in elementary school. The present educational contents about food and nutrition was analysed in the curriculum of elementary school. The results showed the Korean language(20.8%) included an highest ratio in educational contents about food and nutrition, the next was the courses of physical education and wise life(18.1%, each). As the educational contents about food and nutrition in the textbook were dealt with food information (20.8%), Health․Disease(15.3%), and correct dietary habits by order. We could found more contents in the text for the higher classes than for the lower classes. But the most of the contents appeared lack of structure, profundity and continuity for the systematic nutrition education in its entirety. The developed nutrition education contents for discretional activities in this study consist of korean dinning cultures and foreign dinning cultures, correct dinning etiquette, how to choose healthy food, personal sanitary and health, nutrients and food tower, and problem for children’s nutrition as main subject. This six main subjects were composed of 23 subtitles. The teaching manual consisted of the educational goal, background, teaching plan and effect-evaluation plan, and the notice point for the effective lesson. The teaching plan was made for 30 hours and consisted of cooking course, singing/making lyrics, games in nutrition, debate on dietary habit, and role play etc which are oriented to practical learning. We intended to develop this program that attempts to improve in dietary habit of schoolchildren. It is because once formed an adults dietary habit is difficult to change. Schoolchildren’s period is the best adjustable stage. Therefore, nutrition education in elementary stage can change to dietary habit and build the awareness of health.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.4
• /
• pp.28-34
• /
• 2010

In this paper, we proposed an algorithm for utilizing visual information for non-contact predicting method of friction coefficient. Coefficient of friction is very important in driving on road and traversing over obstacle. Our algorithm is based on terrain classification for visual image. The proposed method, non-contacting approach, has advantage over other methods that extract material characteristic of road by sensors contacting road surface. This method is composed of learning group(experiment, grouping material) and predicting friction coefficient group(Bayesian classification prediction function). Every group include previous work of vision. Advantage of our algorithm before entering such terrain can be very useful for avoiding slippery areas. We make experiment on measurement of friction coefficient of terrain. This result is utilized real friction coefficient as prediction method. We show error between real friction coefficient and predicted friction coefficient for performance evaluation of our algorithm.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.4
• /
• pp.2159-2165
• /
• 2013

Cancer registries help to establish and maintain cancer incidence reporting system, serve as a resource for investigation of cancer and its causes, and provide information for planning and evaluation of preventive and control programs. However, their wider role in directly enhancing oncology drug access has not been fully explored. We examined the value of cancer registries in oncology drug access in the Asia-Pacific region on three levels: (1) specific registry variable types; (2) macroscopic strategies on the national level; and (3) a regional cancer registry network. Using literature search and proceedings from an expert forum, this paper covers recent cancer registry developments in eight economies in the Asia-Pacific region - Australia, China, Hong Kong, Malaysia, Singapore, South Korea, Taiwan, and Thailand - and the ways they can contribute to oncology drug access. Specific registry variables relating to demographics, tumor characteristics, initial treatment plans, prognostic markers, risk factors, and mortality help to anticipate drug needs, identify high-priority research area and design access programs. On a national level, linking registry data with clinical, drug safety, financial, or drug utilization databases allows analyses of associations between utilization and outcomes. Concurrent efforts should also be channeled into developing and implementing data integrity and stewardship policies, and providing clear avenues to make data available. Less mature registry systems can employ modeling techniques and ad-hoc surveys while increasing coverage. Beyond local settings, a cancer registry network for the Asia-Pacific region would offer cross-learning and research opportunities that can exert leverage through the experiences and capabilities of a highly diverse region.

• Journal of Distribution Science
• /
• v.12 no.4
• /
• pp.55-63
• /
• 2014

Purpose - Brands and products often play key roles in enabling consumers to experience a good attitude, resulting in mentally enacting a specific prototype and reliving the experience by retelling a specific story. Brand storytelling can function as an important tool for managing the brand. To successfully apply a firm's brand storytelling, it is important to prove the effectiveness of storytelling. Therefore, by utilizing the research of Escalas (1998) and Fog et al. (2005), a list of measurements for storytelling component quality (SCQ) was applied. In addition, customer attitudes toward brand storytelling were tested. In particular, if customers encounter a dynamic and interesting story, although the brand is not widely known, they can be in communion with the brand and establish an emotional connection (Hill, 2003). Thus, brand reputation was divided into two levels (high vs. low), and the difference in effectiveness between storytelling component quality and consumers' advertisement attitude, brand attitude, and purchasing intention was examined. Research design, data, and methodology - By using the measurement list used in Choi, Na, and Hwang (2013), 12 categories in the level of message quality, conflict quality, character quality, and plot quality were measured. In addition, categories of brand reputation, advertisement attitude, brand attitude, and purchasing intention were measured. The study was based on 181 final survey samples targeting undergraduate and graduate students in Gwangju Metropolitan City. Results - Consumer responses toward storytelling were researched in the context of brand characteristics or product attributes, such as brand reputation, differentiated from extant simple effects of storytelling. Some brands with high reputation enjoy a halo effect due to prior learning, while other brands with comparatively low reputation have trouble generating positive responses despite attempts to enhance the level of reputation or induce favorable attitudes. Although not all due to the component quality of storytelling, the case of brands with low reputation exerted more positive impact on consumer attitudes than did brands with high reputation. As mentioned earlier, consumer evaluation of the component quality of storytelling was categorized into advertising attitudes, brand attitudes, and purchase intention for this study; this provides managerial implications in other ways. The results imply that an effective application of storytelling could be an important emotional tool for the development of both brands with low brand awareness and of well-known brands. Finally, this study serves to increase consumers' understanding and ability in interpreting brand stories that marketers tell about themselves, as well as to highlight differential experiences with products by level of brand hierarchy. Conclusions - This research aimed to provide an objective guideline for storytelling component quality while considering brand awareness. Thus, brand reputation was considered for proving the baseline effectiveness of storytelling, and this study provided directions for strategic establishment of storytelling. Based on this, we conclude that in further studies, it will be necessary to systematically manage brand story by considering other situation variables and various story patterns, and studying their differences.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.13 no.1
• /
• pp.93-103
• /
• 2002

The aim of this study is to examine the diagnostic profiles and related clinical variables of children with attention and hyperactivity in psychiatric outpatient clinic. Seventy one children with age range of 5 to 14 were diagnosed by DSM-IV, and assessment battery including KEDI-WISC, KPI-C, ADS(ADHD Diagnostic System) were completed. The subjects were divided into 3 diagnostic groups：ADHD only(n=17), ADHD comorbid(n=27), Other diagnosis(n=27). The results were as follows：In ADHD comorbid group, tic disorder, developmental language disorder, borderline intellectual function, oppositional defiant/conduct disorder, and learning disorder were combined in descending order. Other diagnosis group consisted of tic disorder, borderline intellectual function, depression/anxiety, oppositional defiant/conduct disorder, and others. There were significant differences in IQ, PIQ, and VIQ among the three groups, and ADHD only group showed higher scores of IQ and VIQ than ADHD comorbid group. On the KPI-C, there were no significant differences in all subscales among the three groups. On the visual ADS, omission error and sensitivity showed significant differences among the three groups, and ADHD comorbid group represented higher omission error and lower sensitivity than other diagnostic group. The findings indicated that the inattention and hyperactivity symptoms could be diagnosed into diverse psychiatric disorders in child psychiatry, and ADHD children with comorbidity will show more problems in academic performance and school adjustment.

• Journal of the Korean earth science society
• /
• v.41 no.2
• /
• pp.176-193
• /
• 2020

The purpose of this study was to explore the effects of science classes on positive science experiences and science learner identity, using smart devices in a future school: C middle school. We conducted a paired t test at the beginning and end of the first school year with first-grade students at the future school to investigate positive experiences with science (Shin et al., 2017). Additionally, first and second-grade students in future schools using smart devices wrote and drew their own depictions in science classes to explore science learner identity, based on a modified analytical framework (Luehmann, 2009). The results show that significant effects on science-related career aspirations, self-concepts, and academic emotions were produced by science classes using smart devices. Science classes using smart devices helped students improve their level of agency and activity, solve problems with immediate and sufficient feedback, and experience meaningful perceptions of the nature of science. On the other hand, if students were immature in terms of their use of smart devices, they felt pressured to participate in the classes. The results of this study can be used as a foundation for designing various classroom contexts for the use of smart devices.