• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.177-184
• /
• 2024

In the situation of increasing data at fast speed, we use all kinds of complex ensemble and deep learning algorithms to get the highest accuracy. It's sometimes questionable how these models predict, classify, recognize, and track unknown data. Accomplishing this technique and more has been and would be the goal of intensive research and development in the data science community. A variety of reasons, such as lack of data, imbalanced data, biased data can impact the decision rendered by the learning models. Many models are gaining traction for such interpretations. Now, LIME and SHAP are commonly used, in which are two state of the art open source explainable techniques. However, their outputs represent some different results. In this context, this study introduces a coupling technique of LIME and Shap, and demonstrates analysis possibilities on the decisions made by LightGBM and Keras models in classifying a transaction for fraudulence on the IEEE CIS dataset.

• Journal of Korean Society of Water and Wastewater
• /
• v.37 no.6
• /
• pp.347-361
• /
• 2023

Water utilities are making various efforts to reduce water losses from water networks, and an essential part of them is to recognize the moment when a pipe burst occurs during operation quickly. Several physics-based methods and data-driven analysis are applied using real-time flow and pressure data measured through a SCADA system or smart meters, and methodologies based on machining learning are currently widely studied. Water utilities should apply various approaches together to increase pipe burst detection. The most intuitive and explainable water balance method and its procedure were presented in this study, and the applicability and detection performance were evaluated by applying this approach to water supply pipelines. Based on these results, water utilities can establish a mass balance-based pipe burst detection system, give a guideline for installing new flow meters, and set the detection parameters with expected performance. The performance of the water balance analysis method is affected by the water network operation conditions, the characteristics of the installed flow meter, and event data, so there is a limit to the general use of the results in all sites. Therefore, water utilities should accumulate experience by applying the water balance method in more fields.

• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.523-541
• /
• 2023

As accessibility to 3D printers increases, there is a growing frequency of exposure to chemicals associated with 3D printing. However, research on the toxicity and harmfulness of chemicals generated by 3D printing is insufficient, and the performance of toxicity prediction using in silico techniques is limited due to missing molecular structure data. In this study, quantitative structure-activity relationship (QSAR) model based on data-centric AI approach was developed to predict the toxicity of new 3D printing materials by imputing missing values in molecular descriptors. First, MissForest algorithm was utilized to impute missing values in molecular descriptors of hazardous 3D printing materials. Then, based on four different machine learning models (decision tree, random forest, XGBoost, SVM), a machine learning (ML)-based QSAR model was developed to predict the bioconcentration factor (Log BCF), octanol-air partition coefficient (Log Koa), and partition coefficient (Log P). Furthermore, the reliability of the data-centric QSAR model was validated through the Tree-SHAP (SHapley Additive exPlanations) method, which is one of explainable artificial intelligence (XAI) techniques. The proposed imputation method based on the MissForest enlarged approximately 2.5 times more molecular structure data compared to the existing data. Based on the imputed dataset of molecular descriptor, the developed data-centric QSAR model achieved approximately 73%, 76% and 92% of prediction performance for Log BCF, Log Koa, and Log P, respectively. Lastly, Tree-SHAP analysis demonstrated that the data-centric-based QSAR model achieved high prediction performance for toxicity information by identifying key molecular descriptors highly correlated with toxicity indices. Therefore, the proposed QSAR model based on the data-centric XAI approach can be extended to predict the toxicity of potential pollutants in emerging printing chemicals, chemical process, semiconductor or display process.

• Survey Research
• /
• v.13 no.1
• /
• pp.113-134
• /
• 2012

It is very vital for the presidential preference not only to be a political achievement of the president himself, but also to be an influential base of people-related policy and its implementation. The purpose of this study is to provide the newly developing ideas to compose an appropriate question and to point out the present problems in the utilization of survey results. At first, it turned out to be inappropriate to make a question for evaluation of presidential job approval in the perspective of questionnaire planning. As the respondents were not informed of what the president did, so they were more likely not to know whether the president did well or not. Secondly, the correlation analysis has been made between the evaluation question of the presidential job approval and policy-related questions including the direct one for the presidential preference in the aspect of statistical analysis. Through this process, evaluation of the presidential job approval is not accountable and the direct question for the presidential preference is at best explainable. In conclusion, it seems more persuasive to compare the statistical outcome of the presidential preference which has been derived of to the high rating what the respondents think comprehensively not depending on a subjective decision by mass media.

• Journal of the Korean Magnetics Society
• /
• v.17 no.2
• /
• pp.51-54
• /
• 2007

By substituting Cr in inverse-spinel $Fe_3O_4,\;Cr_xFe_{3-x}O_4$ thin film samples were prepared by sol-gel spin-coating method and their structural electronic, and magnetic properties were analyzed. X-ray diffraction indicates that the lattice constant decrease with increasing Cr composition (x). This result can be explained in terms of occupation of octahedral sites by $Cr^{3+}$ ions with smaller ionic radius than that of $Fe^{3+}$ Vibrating sample magnetometry measurements on the samples at room temperature revealed that saturation magnetization ($M_s$) decrease by Cr substitution, explainable by comparing spin magnetic moment among the related transition-metal ions. A decrease of magnetoresistence effect with x was observed, similar to that of $M_s$. The coercivity of the $Cr_xFe_{3-x}O_4$ films was found to increase with x, attributed to the increase of magnetic anisotropy by the existence of octahedral $Cr^{3+}(d^3)$.

• Journal of Korean Home Economics Education Association
• /
• v.9 no.1
• /
• pp.1-17
• /
• 1997

This study was conducted to obtain basic materials to improve the teaching method of Home Economics by theoretically looking into the supplementary teaching materials or implements usable in teaching Costume History area. And based on these data, the types and the applications of the supplementary teaching materials or implements highschool owned were examined. The subjects of this study were 111 Home Economics and Housework curriculum highschool teachers who give a lecture in the country by using self-administered questionnaires. SAS program was used to calculate frequency, percentage, average, standard deviation, and $\chi$(sup)2-test analysis. The results of the study were as follows； 1. Most of the highschool teachers used the school expenses for experiments in preparing the supplementary teaching materials or implements. 2. Of the supplementary teaching materials and implements concerning Costume History, visual implements such as slides and pictures were the mostly owned. CD and audio implements as cassette-tapes were not used. 3. Most of the teachers recognized the importance of the audio-visual teaching materials and implements concerning Costume History. 4. Among the audio-visual materials and implements concerning Costume History by which can be made by school teachers of Home Economics and Housework curriculum, the mostly used one was ‘cutting pictorials from magazines and newspapers’, and the next were ‘orbital materials’, and ‘copy the pictorials’, and the least was ‘recording from the radio’. 5. Most of the annual expenses assigned to the department of Home Economics was used in cooking practice, and the least of the expenses was assigned in buying audio-visual teaching materials and implements. 6. Time assigned to the area of Home Economics was for the most part one or two hours per week, and among this, time assigned to the history of western costume and the history ok korean costume was for the most part five to eight hours. 7. The areas that the highschool teachers felt difficulties mostly during clothing and textiles curriculum were ‘textiles’and the next were ‘knitting’, ‘western costume history’, and ‘korean clothing construction’. 8. The difficulties the highschool teachers faced while teaching Costume History were mostly that ‘the pictorials in the text is not fully explainable’, the next were ‘most of the supplementary teaching materials or implements are not owned’, ‘have to explain very much in a short time’, and ‘the lectural explanation is insufficient’. 9. The solution for the difficulties that the highschool teachers faced while teaching Costume History was mostly ‘the information, on which audio-visual materials and implements are distributed in the market, should be easy to obtain’, the next opinions were ‘the school should provide enough experiment and practice expenses to buy audio-visual materials and implements’, and ‘education facilities of the Home Economics Department should be the main aspects in improving the teaching methods and should give special lectures about it’.

• Journal of Preventive Medicine and Public Health
• /
• v.26 no.1 s.41
• /
• pp.74-85
• /
• 1993

To measure the serum succinyl trialanine p-nitroanilide hydrolytic activity as new index of liver function in workers exposed to organic solvents, this study conducted 114 workers in department of shoe-making of shoes factories. The results obtained from this study were as follows : 1. The mean values of serum GOT, GPT, ${\gamma}GT$ in whole workers were $22{\pm}12.32,\;20{\pm}9.05,\;28{\pm}21.35IU/l$, respectively and the mean value of serum STN hydrolytic activity was $0.08{\pm}0.05$. 2. The serum STN hydrolytic activity was significantly higher for male (p<0.05) and there was no difference among the groups of age. 3. There was no difference in the groups by working hours but significant difference in persons who worked over 3 years or were exposed to toluene over 100ppm (p<0.05). 4. The correlation of the exposed dose of toluene and serum GOT, GPT, ${\gamma}GT$ and serum STN hydrolytic activity were statistically significant (r=0.027-0.518). 5. The exposed dose of toluene was most explainable variable and statistically significant among the factors affecting serum STN hydrolytic activity (p<0.05).

• Journal of the Korean Vacuum Society
• /
• v.16 no.4
• /
• pp.262-266
• /
• 2007

Polycrystalline $Mn_xFe_{3-x}O_4$ thin films were synthesized on Si(100) substrates using sol-gel method and the effects of Mn substitution on the structural, magnetic, and magnetotransport properties were analyzed. X-ray diffraction revealed that cubic structure is maintained up to x = 1.78 with increasing lattice constant for increasing x. Such increase of the lattice constant is attributable to the substitution of $Mn^{2+}$ (with larger ionic radius) ions into tetrahedral $Fe^{3+}$(with smaller ionic radius) sites. VSM measurements revealed that $M_s$ does not vary significantly with x, qualitatively explainable by comparing spin magnetic moments of Mn and Fe ions. On the other hand, $H_c$ was found to decrease with increasing x, attributable to the decrease of magnetic anisotropy due to the decrease of $Fe^{2+}$ density through $Mn^{2+}$ substitution. Magnetoresistance (MR) of the $Mn_xFe_{3-x}O_4$ films was found to decrease with increasing x. Analysis of the MR data in comparison with the VSM results gives an indication of the tunneling of spin-polarized carriers through the grain boundaries of the polycrystalline samples at low external field and spin-flip of the carriers at high external field.

• Proceedings of the Korean Association for Survey Research Conference
• /
• 2011.10a
• /
• pp.63-81
• /
• 2011

The purpose of this study is to provide the newly developing ideas to compose an appropriate question and to point out the present problems in the utilization of survey result. It is very vital for the presidential preference not only to be a political achievement of the President himself, but also to be a influential base of people-related policy and its implementation. At first, it turned out to be inappropriate to make a question for the evaluation of presidential working performance in the perspective of questionnaire planning. As the respondents were not informed of what the President did, so they were more likely not to know whether the President did well or not. Secondly, the correlation analysis has been made between the evaluation question of the presidential working performance and policy-related questions including the direct one for the presidential preference in the aspect of statistical analysis. Through this process, the evaluation of the presidential working performance is not accountable and the direct question for the presidential preference is at best explainable. In conclusion, it seems more persuasive to compare the statistical outcome of the presidential preference which has been derived of to the high rating what the respondents think comprehensively not depending on a subjective decision by mass media.

• The Journal of the Convergence on Culture Technology
• /
• v.6 no.4
• /
• pp.789-795
• /
• 2020

Many deaths have been reported due to the worldwide pandemic of COVID-19. In order to prevent the further spread of COVID-19, it is necessary to quickly and accurately read images of suspected patients and take appropriate measures. To this end, this paper introduces a deep learning-based COVID-19 chest X-ray reading technique that can assist in image reading by providing medical staff whether a patient is infected. First of all, in order to learn the reading model, a sufficient dataset must be secured, but the currently provided COVID-19 open dataset does not have enough image data to ensure the accuracy of learning. Therefore, we solved the image data number imbalance problem that degrades AI learning performance by using a Stacked Generative Adversarial Network(StackGAN++). Next, the DenseNet-based classification model was trained using the augmented data set to develop the reading model. This classification model is a model for binary classification of normal chest X-ray and COVID-19 chest X-ray, and the performance of the model was evaluated using part of the actual image data as test data. Finally, the reliability of the model was secured by presenting the basis for judging the presence or absence of disease in the input image using Grad-CAM, one of the explainable artificial intelligence called XAI.