• Journal of Korea Water Resources Association
• /
• v.56 no.11
• /
• pp.801-813
• /
• 2023

Recently, as studies about Artificial Neural Network (ANN) are actively progressing, studies for predicting water quality of rivers using ANN are being conducted. However, it is difficult to analyze the operation process inside ANN, because ANN is form of Black-box. Although eXplainable Artificial Intelligence (XAI) is used to analyze the computational process of ANN, research using XAI technology in the field of water resources is insufficient. This study analyzed Multi Layer Perceptron (MLP) to predict Water Temperature (WT), Dissolved Oxygen (DO), hydrogen ion concentration (pH) and Chlorophyll-a (Chl-a) at the Dasan water quality observatory in the Nakdong river using Layer-wise Relevance Propagation (LRP) among XAI technologies. The MLP that learned water quality was analyzed using LRP to select the optimal input data to predict water quality, and the prediction results of the MLP learned using the optimal input data were analyzed. As a result of selecting the optimal input data using LRP, the prediction accuracy of MLP, which learned the input data except daily precipitation in the surrounding area, was the highest. Looking at the analysis of MLP's DO prediction results, it was analyzed that the pH and DO a had large influence at the highest point, and the effect of WT was large at the lowest point.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.3
• /
• pp.160-165
• /
• 2023

In this study, we propose an SNS sentence writing assistance system that utilizes YOLO and GPT to assist users in writing texts with images, such as SNS. We utilize the YOLO model to extract objects from images inserted during writing, and also extract meta-information such as GPS information and creation time information, and use them as prompt values for GPT. To use the YOLO model, we trained it on form image data, and the mAP score of the model is about 0.25 on average. GPT was trained on 1,000 blog text data with the topic of 'restaurant reviews', and the model trained in this study was used to generate sentences with two types of keywords extracted from the images. A survey was conducted to evaluate the practicality of the generated sentences, and a closed-ended survey was conducted to clearly analyze the survey results. There were three evaluation items for the questionnaire by providing the inserted image and keyword sentences. The results showed that the keywords in the images generated meaningful sentences. Through this study, we found that the accuracy of image-based sentence generation depends on the relationship between image keywords and GPT learning contents.

• Korean Journal of Cognitive Science
• /
• v.29 no.2
• /
• pp.121-135
• /
• 2018

Dancers' motions are perceived by observers through visual processes with visual information forming the basis for the observers' appreciation and evaluation of the dancers' motions. There have been many discussions as to whether or not observers' personal athletic capabilities form an essential basis for accurate assessment of the motions of others but, so far, no valid conclusions have been reached. The purpose of this study is to investigate how the ability to predict motions of others varies depending on the athletic expertise of the observers. Participants of this research were ballet dancers of varying athletic expertise. Twenty seven participants were divided into three groups with nine in each: beginners, intermediate experts and experts. The participants watched the same dance video and then evaluated whether the motion would be successful or not. The movement related visual information required to evaluate the success of the motion was systematically adjusted by controlling the length of the video. Using the temporal occlusion method, this study measured the response accuracy of the participants by category of expertise. Under the circumstance with insufficient visual information to utilize, the experts showed higher rates of correct response than the intermediate experts and the beginners. The beginners showed higher rates of wrong response than the experts and the intermediate experts. These results showed that the ability to predict success or failure of a dance motion varied depending on motion expertise of the observers, although they had similar level of expertise in perception. Participants considered to have high athletic expertise showed high prediction ability on the result of the motion. In addition, high expertise in perception reduced the likelihood that participants would make hasty responses under the circumstance with insufficient information and helped to reduce wrong response rate. In conclusion, this study showed that motor expertise and perceptual expertise contribute to prediction accuracy of observed motions.

• Korean Journal of Oriental Medicine
• /
• v.3 no.1
• /
• pp.1-15
• /
• 1997

${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$(1445) is a book compiled the medical achievements of China and Choseon in those times and it's our source of pride to have it In this country. It also deserves careful investigation since this book can provide some clues of features of missing books in China and Korea. The extent of accuracy of xylographica of old books determines the possiblity of in depth further study. So authors attempted to investigate the xylographica of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ one of the 3 main books in Korea. Previous investigation done by Miki Sakae and Kim Doo Jong are noticeable. On the basis of their respective works, we analyzed 'Annals of the Choseon Dynasty' to find records related with ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ and estimated the situation of its publication. We tried figure the situation of those times of China, Japan and Korea(including North Korea) and tried to estimate the book's original xylographica as much as we could. By King Sejong's command, the first draft of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ consisted of 365 books was made by collaboration of civil officials and medical officers during the period from 1443 to 1445. And then from 1451(first year of Moonjong's reign) to 1464(l0th year of Sejo's reign) lots of manpowers were employed and through the process of countless erasure, proofreading, arrangement and rearrangement revised version of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ which is called by Sejo text was completed. After 3 years of wood engraving work, the first printed form of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ (alternately called Seongjong text) in folding case consisted of 266 chapters, 264 volumes came into the world in 1477.(8th year of Seongjong's reign). This was 32 years after the initial completion of the edition. So ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ exists in three forms as Sejong text, Sejo text and Seongjong text respectively. Since those texts were plundered during the Japanese invasion of Korea in 1592, none of the original copy remains within korea. The texts were constantly moved to kadeungcheongieong, to Kongdeungpyeongio, Jesookoan of Edo, to East University of department of classic books, to Cheoncho archives, to the Imperial Museum and finally is kept in the royal palace at present. (Doseoryo text Eulhae printing type) Reduced-size republication books of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ in wooden type were imported at the time of 'Byeongja Korea-Japan Treaty in 1876' and of those 2 books, one copy was treasured in the Royal Household of the Yi Dynasty and than was lost during the Korean War circa 1950. The other remaining copy has been kept succesively by Kojong's imperial grant, Royal doctor Hong Cheol Bo, Hong Taek Joo, Hong Ik Pyo the book agent, and now is kept In Yonsei University Library and this is the only existing copy in Korea at present. In 1965, Dongyang Medical college published the transcription version of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ consisting of 11 books and then in 1981 after edition and arrangement by Choonghoa(中華) publishing company, photoprint copy of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ was published in Keumgang(金剛) publishing company In 1991, October Yeokang(驛江) publishing company producd photocopies of ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ which were previously translated into Korean by North Korea Institute of Oriental Medicine and then issued by medical publishing company. In China, two institutes, Zhejiang Institute of Traditional Chinese Medicine and Huzhou Traditional Chinese Medical Hospital cooperated to publish a revised and marked text consiting of 11 books by adding marking points to japanse Edohakhoondang text which were used as a reference. Both the korean and chinese texts issued were grounded by the ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ kept in the royal palace. Any further study concerning ${\ulcorner}$Classified Collection of Medical Prescriptions${\lrcorner}$ can acquire its accuracy and objectivity when the japanese text kept in the royal palace is taken as an original copy.

• The Korean Journal of Nuclear Medicine
• /
• v.30 no.4
• /
• pp.548-559
• /
• 1996

A series of performance measurements of positron emission tomography (PET) were performed following the recommendations of the Computer and Instrumentation Council of the Society of Nuclear Medicine and the National Electrical Manufacturers Association. We investigated the performance of the General Electric $Advance^{TM}$ PET. The measurements include the basic intrinsic tests of spatial resolution, scatter fraction, sensitivity, and count rate losses and randoms. They also include the tests of the accuracy of corrections: count rate linearity correction, uniformity correction, scatter correction and attenuation correction. GE $Advance^{TM}$ PET has bismuth germanate oxide crystals (4.0mm transaxial ${\times}$ 8.1mm axial ${\times}$ 30.0mm radial) in 18 rings, which form 35 imaging planes spaced by 4.25mm. The system has retractable tungsten septa 1mm thick and 12cm long. Transaxial resolution was 4.92mm FWHM in 2D and 5.14mm FWHM in 3D at the center. Average axial resolution in 2D decreased from 3.91mm FWHM at the center to 6.49mm FWHM at R=20cm. Average scatter fraction of direct and cross slices was 9.57%. Dead-time losses of 50% corresponded to a radioactivity concentration of $4.86{\mu}Ci/cc$ and a true count rate of 519 kcps in 2D. The accuracy of count rate linearity correction was 1.84% at the activity of $4.50{\mu}Ci/cc$. Non-uniformity was 2.06% in 2D and 2.93% in 3D. Remnant errors after scatter correction were 0.55% in 2D and 4.12% in 3D. The errors of attenuation correction were 6.21% (air), 0.20% (water), -6.32% (teflon) in 2D and 5.00% (air), 6.94% (water), 3.01% (teflon) in 3D. The results indicate the performance of GE $Advance^{TM}$ PET scanner to be well suited for clinical and research applications.

• Journal of Wetlands Research
• /
• v.12 no.1
• /
• pp.95-103
• /
• 2010

Discharge data examine the process of hydrologic cycle and used significantly in water resource planning and irrigation and flood control planning. However, it needs lots of time and money to get the discharge data. So discharge rating curve is usually used in converting discharge data. Therefore reliability of discharge rating curve absolutely depends on quality of discharge data. Many engineers who study hydrologic engineering make high quality discharge data to develop reliable discharge rating curve. And they carry out research on standard and method of discharge measurement, and equipment improvement. Now various flow meters are utilized to make discharge data in Korea. However, accuracy of equipment and experimental research data from measurement are not enough. In this paper, constant discharge flowed through standard concrete channel, and the velocity is measured using various flow meters. Also Discharge is calculated by measured data to compare and analyze. The equipment for the experiment is Price AA(USGS Type AA Current meter), flow meter, ADC, C2 small current meter, flow tracker, Electromagnetic current meter. The discharge got form various flow meters which are widely used for discharge measurement. The various depths of water were examined and compared such as 0.30 m, 0.35 m, 0.40 m, 0.45 m, 0.50 m, 0.55 m. The experiment progresses a round-measurement on 6-case. Wading measurement(one point method : the 60 % height in surface of the water) was applied to improve creditability and accuracy among measurement methods. USGS Type AA current Meter, Flow Meter, ADC, C2 Small Current meter got the certificate of quality guaranteed. So the results of experiment were used to compare discharge. The Results showed the difference based on USGS Type AA current Meter at average discharge and velocity. Electromagnetic current meter made differences over $\pm$ 10 % and Flow Meter made differences under $\pm$ 10 %. Also ADC, Flow Meter, C2 Small Current meter made differences under $\pm$ 5 %.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.27-42
• /
• 2020

Traditional companies with offline stores were unable to secure large display space due to the problems of cost. This limitation inevitably allowed limited kinds of products to be displayed on the shelves, which resulted in consumers being deprived of the opportunity to experience various items. Taking advantage of the virtual space called the Internet, online shopping goes beyond the limits of limitations in physical space of offline shopping and is now able to display numerous products on web pages that can satisfy consumers with a variety of needs. Paradoxically, however, this can also cause consumers to experience the difficulty of comparing and evaluating too many alternatives in their purchase decision-making process. As an effort to address this side effect, various kinds of consumer's purchase decision support systems have been studied, such as keyword-based item search service and recommender systems. These systems can reduce search time for items, prevent consumer from leaving while browsing, and contribute to the seller's increased sales. Among those systems, recommender systems based on association rule mining techniques can effectively detect interrelated products from transaction data such as orders. The association between products obtained by statistical analysis provides clues to predicting how interested consumers will be in another product. However, since its algorithm is based on the number of transactions, products not sold enough so far in the early days of launch may not be included in the list of recommendations even though they are highly likely to be sold. Such missing items may not have sufficient opportunities to be exposed to consumers to record sufficient sales, and then fall into a vicious cycle of a vicious cycle of declining sales and omission in the recommendation list. This situation is an inevitable outcome in situations in which recommendations are made based on past transaction histories, rather than on determining potential future sales possibilities. This study started with the idea that reflecting the means by which this potential possibility can be identified indirectly would help to select highly recommended products. In the light of the fact that the attributes of a product affect the consumer's purchasing decisions, this study was conducted to reflect them in the recommender systems. In other words, consumers who visit a product page have shown interest in the attributes of the product and would be also interested in other products with the same attributes. On such assumption, based on these attributes, the recommender system can select recommended products that can show a higher acceptance rate. Given that a category is one of the main attributes of a product, it can be a good indicator of not only direct associations between two items but also potential associations that have yet to be revealed. Based on this idea, the study devised a recommender system that reflects not only associations between products but also categories. Through regression analysis, two kinds of associations were combined to form a model that could predict the hit rate of recommendation. To evaluate the performance of the proposed model, another regression model was also developed based only on associations between products. Comparative experiments were designed to be similar to the environment in which products are actually recommended in online shopping malls. First, the association rules for all possible combinations of antecedent and consequent items were generated from the order data. Then, hit rates for each of the associated rules were predicted from the support and confidence that are calculated by each of the models. The comparative experiments using order data collected from an online shopping mall show that the recommendation accuracy can be improved by further reflecting not only the association between products but also categories in the recommendation of related products. The proposed model showed a 2 to 3 percent improvement in hit rates compared to the existing model. From a practical point of view, it is expected to have a positive effect on improving consumers' purchasing satisfaction and increasing sellers' sales.

• Korean Journal of Clinical Laboratory Science
• /
• v.37 no.2
• /
• pp.65-70
• /
• 2005

Those specifications of the CLIA analytical tolerance limits are consistent with the performance goals in Six Sigma Quality Management. Six sigma analysis determines performance quality from bias and precision statistics. It also shows if the method meets the criteria for the six sigma performance. Performance standards calculates allowable total error from several different criteria. Six sigma means six standard deviations from the target value or mean value and about 3.4 failures per million opportunities for failure. Sigma Quality Level is an indicator of process centering and process variation total error allowable. Tolerance specification is replaced by a Total Error specification, which is a common form of a quality specification for a laboratory test. The CLIA criteria for acceptable performance in proficiency testing events are given in the form of an allowable total error, TEa. Thus there is a published list of TEa specifications for regulated analytes. In terms of TEa, Six Sigma Quality Management sets a precision goal of TEa/6 and an accuracy goal of 1.5 (TEa/6). This concept is based on the proficiency testing specification of target value +/-3s, TEa from reference intervals, biological variation, and peer group median mean surveys. We have found rules to calculate as a fraction of a reference interval and peer group median mean surveys. We studied to develop total error allowable from peer group survey results and CLIA 88 rules in US on 19 items TP, ALB, T.B, ALP, AST, ALT, CL, LD, K, Na, CRE, BUN, T.C, GLU, GGT, CA, phosphorus, UA, TG tests in chematology were follows. Sigma level versus TEa from peer group median mean CV of each item by group mean were assessed by process performance, fitting within six sigma tolerance limits were TP ($6.1{\delta}$/9.3%), ALB ($6.9{\delta}$/11.3%), T.B ($3.4{\delta}$/25.6%), ALP ($6.8{\delta}$/31.5%), AST ($4.5{\delta}$/16.8%), ALT ($1.6{\delta}$/19.3%), CL ($4.6{\delta}$/8.4%), LD ($11.5{\delta}$/20.07%), K ($2.5{\delta}$/0.39mmol/L), Na ($3.6{\delta}$/6.87mmol/L), CRE ($9.9{\delta}$/21.8%), BUN ($4.3{\delta}$/13.3%), UA ($5.9{\delta}$/11.5%), T.C ($2.2{\delta}$/10.7%), GLU ($4.8{\delta}$/10.2%), GGT ($7.5{\delta}$/27.3%), CA ($5.5{\delta}$/0.87mmol/L), IP ($8.5{\delta}$/13.17%), TG ($9.6{\delta}$/17.7%). Peer group survey median CV in Korean External Assessment greater than CLIA criteria were CL (8.45%/5%), BUN (13.3%/9%), CRE (21.8%/15%), T.B (25.6%/20%), and Na (6.87mmol/L/4mmol/L). Peer group survey median CV less than it were as TP (9.3%/10%), AST (16.8%/20%), ALT (19.3%/20%), K (0.39mmol/L/0.5mmol/L), UA (11.5%/17%), Ca (0.87mg/dL1mg/L), TG (17.7%/25%). TEa in 17 items were same one in 14 items with 82.35%. We found out the truth on increasing sigma level due to increased total error allowable, and were sure that the goal of setting total error allowable would affect the evaluation of sigma metrics in the process, if sustaining the same process.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.36 no.3
• /
• pp.128-136
• /
• 2018

This study was centered on the protested data of the issue of "History of Landscape Architecture" in the handwritten course of landscaping articles of National Qualifications Test. The purpose of this study is to examine the types of social problems in the process of correcting erroneous historical facts. The purpose of this study was to find alternatives for the development of the field of landscape and culture history that can assist in the verification of the historical facts of the landscape sciences examination questions. The main results are as follows. First, as a result of analyzing the contents of the landscape architects' subject matter, the establishment of concept of landscape style and form and the confirmation of historical facts were investigated as important types to be established for development of landscape landscape history department. It seems that the social consensus of the expert group is needed to supplement the lack of data to refer to landscape architectural theory. Second, the analysis of the problematic narrative contents resulted in a total of five types of questionnaires. The appeared in the Undefined style and form(52.94%), Unproven historical facts(25.13%), Obscurity Era classification(11.77%), Lack of specificity(6.95%), Content scope of obscurity events(3.21%) Third, it is not only the lack of information to learn the theory by comparing and analyzing the contents of the statements in the landscape architect 's question items, but also the difference of contents between books was analyzed as the main cause of the problem. As a result of examining the characteristics and examples of the issues raised in landscape architectural problems, it was related to the social phenomenon, and it was classified into cultural factors and political factors. Fourth, the resolution of problematic issues in landscape architects' landscaping articles, which are national technical qualification tests, shows positive results. The information determined in the process of solving the perceived content can be used directly in landscaping field, and it helps the accuracy of the verification process by identifying the types and characteristics of the issues.