• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• Journal of Intelligence and Information Systems
• /
• v.27 no.4
• /
• pp.1-22
• /
• 2021

Aspect Based Sentiment Analysis (ABSA), which analyzes sentiment based on aspects that appear in the text, is drawing attention because it can be used in various business industries. ABSA is a study that analyzes sentiment by aspects for multiple aspects that a text has. It is being studied in various forms depending on the purpose, such as analyzing all targets or just aspects and sentiments. Here, the aspect refers to the property of a target, and the target refers to the text that causes the sentiment. For example, for restaurant reviews, you could set the aspect into food taste, food price, quality of service, mood of the restaurant, etc. Also, if there is a review that says, "The pasta was delicious, but the salad was not," the words "steak" and "salad," which are directly mentioned in the sentence, become the "target." So far, in ABSA, most studies have analyzed sentiment only based on aspects or targets. However, even with the same aspects or targets, sentiment analysis may be inaccurate. Instances would be when aspects or sentiment are divided or when sentiment exists without a target. For example, sentences like, "Pizza and the salad were good, but the steak was disappointing." Although the aspect of this sentence is limited to "food," conflicting sentiments coexist. In addition, in the case of sentences such as "Shrimp was delicious, but the price was extravagant," although the target here is "shrimp," there are opposite sentiments coexisting that are dependent on the aspect. Finally, in sentences like "The food arrived too late and is cold now." there is no target (NULL), but it transmits a negative sentiment toward the aspect "service." Like this, failure to consider both aspects and targets - when sentiment or aspect is divided or when sentiment exists without a target - creates a dual dependency problem. To address this problem, this research analyzes sentiment by considering both aspects and targets (Target-Aspect-Sentiment Detection, hereby TASD). This study detected the limitations of existing research in the field of TASD: local contexts are not fully captured, and the number of epochs and batch size dramatically lowers the F1-score. The current model excels in spotting overall context and relations between each word. However, it struggles with phrases in the local context and is relatively slow when learning. Therefore, this study tries to improve the model's performance. To achieve the objective of this research, we additionally used auxiliary loss in aspect-sentiment classification by constructing CNN(Convolutional Neural Network) layers parallel to existing models. If existing models have analyzed aspect-sentiment through BERT encoding, Pooler, and Linear layers, this research added CNN layer-adaptive average pooling to existing models, and learning was progressed by adding additional loss values for aspect-sentiment to existing loss. In other words, when learning, the auxiliary loss, computed through CNN layers, allowed the local context to be captured more fitted. After learning, the model is designed to do aspect-sentiment analysis through the existing method. To evaluate the performance of this model, two datasets, SemEval-2015 task 12 and SemEval-2016 task 5, were used and the f1-score increased compared to the existing models. When the batch was 8 and epoch was 5, the difference was largest between the F1-score of existing models and this study with 29 and 45, respectively. Even when batch and epoch were adjusted, the F1-scores were higher than the existing models. It can be said that even when the batch and epoch numbers were small, they can be learned effectively compared to the existing models. Therefore, it can be useful in situations where resources are limited. Through this study, aspect-based sentiments can be more accurately analyzed. Through various uses in business, such as development or establishing marketing strategies, both consumers and sellers will be able to make efficient decisions. In addition, it is believed that the model can be fully learned and utilized by small businesses, those that do not have much data, given that they use a pre-training model and recorded a relatively high F1-score even with limited resources.

• Journal of Intelligence and Information Systems
• /
• v.23 no.1
• /
• pp.95-108
• /
• 2017

Recently, AlphaGo which is Bakuk (Go) artificial intelligence program by Google DeepMind, had a huge victory against Lee Sedol. Many people thought that machines would not be able to win a man in Go games because the number of paths to make a one move is more than the number of atoms in the universe unlike chess, but the result was the opposite to what people predicted. After the match, artificial intelligence technology was focused as a core technology of the fourth industrial revolution and attracted attentions from various application domains. Especially, deep learning technique have been attracted as a core artificial intelligence technology used in the AlphaGo algorithm. The deep learning technique is already being applied to many problems. Especially, it shows good performance in image recognition field. In addition, it shows good performance in high dimensional data area such as voice, image and natural language, which was difficult to get good performance using existing machine learning techniques. However, in contrast, it is difficult to find deep leaning researches on traditional business data and structured data analysis. In this study, we tried to find out whether the deep learning techniques have been studied so far can be used not only for the recognition of high dimensional data but also for the binary classification problem of traditional business data analysis such as customer churn analysis, marketing response prediction, and default prediction. And we compare the performance of the deep learning techniques with that of traditional artificial neural network models. The experimental data in the paper is the telemarketing response data of a bank in Portugal. It has input variables such as age, occupation, loan status, and the number of previous telemarketing and has a binary target variable that records whether the customer intends to open an account or not. In this study, to evaluate the possibility of utilization of deep learning algorithms and techniques in binary classification problem, we compared the performance of various models using CNN, LSTM algorithm and dropout, which are widely used algorithms and techniques in deep learning, with that of MLP models which is a traditional artificial neural network model. However, since all the network design alternatives can not be tested due to the nature of the artificial neural network, the experiment was conducted based on restricted settings on the number of hidden layers, the number of neurons in the hidden layer, the number of output data (filters), and the application conditions of the dropout technique. The F1 Score was used to evaluate the performance of models to show how well the models work to classify the interesting class instead of the overall accuracy. The detail methods for applying each deep learning technique in the experiment is as follows. The CNN algorithm is a method that reads adjacent values from a specific value and recognizes the features, but it does not matter how close the distance of each business data field is because each field is usually independent. In this experiment, we set the filter size of the CNN algorithm as the number of fields to learn the whole characteristics of the data at once, and added a hidden layer to make decision based on the additional features. For the model having two LSTM layers, the input direction of the second layer is put in reversed position with first layer in order to reduce the influence from the position of each field. In the case of the dropout technique, we set the neurons to disappear with a probability of 0.5 for each hidden layer. The experimental results show that the predicted model with the highest F1 score was the CNN model using the dropout technique, and the next best model was the MLP model with two hidden layers using the dropout technique. In this study, we were able to get some findings as the experiment had proceeded. First, models using dropout techniques have a slightly more conservative prediction than those without dropout techniques, and it generally shows better performance in classification. Second, CNN models show better classification performance than MLP models. This is interesting because it has shown good performance in binary classification problems which it rarely have been applied to, as well as in the fields where it's effectiveness has been proven. Third, the LSTM algorithm seems to be unsuitable for binary classification problems because the training time is too long compared to the performance improvement. From these results, we can confirm that some of the deep learning algorithms can be applied to solve business binary classification problems.

• Korean Journal of Remote Sensing
• /
• v.29 no.5
• /
• pp.517-525
• /
• 2013

Time-series data of Normal Difference Vegetation Index (NDVI) obtained by the Moderate-resolution Imaging Spectroradiometer(MODIS) satellite imagery gives a waveform that reveals the characteristics of the phenology. The waveform can be decomposed into harmonics of various periods by the Fourier transformation. The resulting $n^{th}$ harmonics represent the amount of NDVI change in a period of a year divided by n. The values of each harmonics or their relative relation have been used to classify the vegetation species and to build a vegetation map. Here, we propose a method to estimate the annual amount of carbon absorbed on the forest from the $1^{st}$ harmonic NDVI value. The $1^{st}$ harmonic value represents the amount of growth of the leaves. By the allometric equation of trees, the growth of leaves can be considered to be proportional to the total amount of carbon absorption. We compared the $1^{st}$ harmonic NDVI values of the 6220 sample points with the reference data of the carbon absorption obtained by the field survey in the forest of South Korea. The $1^{st}$ harmonic values were roughly proportional to the amount of carbon absorption irrespective of the species and ages of the vegetation. The resulting proportionality constant between the carbon absorption and the $1^{st}$ harmonic value was 236 tCO2/5.29ha/year. The total amount of carbon dioxide absorption in the forest of South Korea over the last ten years has been estimated to be about 56 million ton, and this coincides with the previous reports obtained by other methods. Considering that the amount of the carbon absorption becomes a kind of currency like carbon credit, our method is very useful due to its generality.

• Korean Chemical Engineering Research
• /
• v.54 no.4
• /
• pp.533-542
• /
• 2016

We investigate fluidization characteristics of the mixture of rice husk, silica sand and rice husk ash as a preliminary study for valuable utilization of rice husk ash obtained from gasification of rice husk in a fluidized bed reactor. As experiment valuables, the blending ratio of rice husk and sand (rice husk: sand) is selected as 5:95, 10:90, 20:80 and 30:70 on a volume base. Rice husk ash was added with 6 vol% of rice husk for each experiment and air velocity to the reactor was 0~0.63 m/s. In both rice husk/sand and rice husk/sand/ash mixture, the minimum fluidization velocity (Umf) is observed as 0.19~0.21 m/s at feeding of 0~10 vol.% of rice husk and 0.30 m/s at feeding of 20 vol.% of rice husk. With increasing the amount of rice husk up to 30 vol.%, $U_{mf}$ can not measure due to segregation behavior. The mixing index for each experiment is determined using mixing index equation proposed by Brereton and Grace. The mixing index of the mixture of rice husk/sand and rice husk/sand/ash was 0.8~1 and 0.88~1, respectively. The optimum fluidization condition was found for the good mixing and separation of rice husk ash.

• Food Science and Preservation
• /
• v.24 no.8
• /
• pp.1103-1112
• /
• 2017

This study was performed to investigate the effects of different cooking methods (boiling, roasting, stir-frying, and deep-frying) on folate retention in 6 kinds of mushrooms (Beech-, button-, Juda's ear-, oak-, oyster-, and winter-mushrooms) frequently consumed in Korea. In order to assure reliability of analytical data, trienzyme extraction-L casei method was verified and analytical quality control was also evaluated. Folate contents of mushrooms varied by 6.04-64.82 g/100 g depending on the type of mushrooms. and were significantly affected by cooking methods. Depending on cooking methods, folate contents of mushrooms decreased by 22-48%, 2-31%, and 17-56% for Juda's ear-, oak- and oyster-mushrooms, respectively, while 17-90% of folate was increased in Beech mushroom. Overall, the largest weight loss was found in boiled mushrooms, but the lowest one in deep-fried samples. True folate retention rates considering processing factor were less than 100% for all cooked mushrooms except for Beech samples. Overall, folate loss was the largest by boiling with water but the smallest by deep-frying. Both accuracy and precision of trienzyme extraction-L-casei method were excellent based on a recovery close to 100% and coefficient variations less than 3%. Quality control chart of folate analysis (n=26) obtained during the entire study and an international proficiency test (z-score=-0.5) showed that trienzyme extraction-L casei method is reliable enough for production of national folate database.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.46 no.11
• /
• pp.1386-1396
• /
• 2017

The purpose of this study was to determine the optimum Platycodon grandiflorum root concentrate (PGRC, $65^{\circ}Brix$), fermented P. grandiflorum root extract by Lactobacillus plantarum (FPGRE, $2^{\circ}Brix$), and cactus Chounnyouncho extract (Cactus-E, $2^{\circ}Brix$) for preparation of PGRC stick product with FPGRE using response surface methodology (RSM). The experimental conditions were designed according to a central composite design with 20 experimental points, including three replicates for three independent variables such as amount of PGRC (8~12 g), FPGRE (0~20 g), and Cactus-E (0~20 g). The experimental data for the sensory evaluation and functional properties based on antioxidant activity and antimicrobial activity were fitted with the quadratic model, and accuracy of equations was analyzed by ANOVA. For the responses, sensory and functional properties showed significant correlation with contents of three independent variables. The results indicate that addition of PGRC contributed to increased bitterness and acridity based on the sensory test and antimicrobial activity, addition of FPGRE contributed to increased antioxidant activity and antimicrobial activity, and addition of Cactus-E contributed to increased fluidity based on the sensory test, antioxidant activity, and antimicrobial activity. Based on the results of RSM, the optimum formulation of PGRC stick product was calculated as PGRC 8.456 g, FPGRE 20.00 g, and Cactus-Ex 20.00 g with minimal bitterness and acridity, as well as optimized fluidity, antioxidant activity, and antimicrobial activity.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1505-1514
• /
• 2022

Image matching is a crucial preprocessing step for effective utilization of multi-temporal and multi-sensor very high resolution (VHR) satellite images. Deep learning (DL) method which is attracting widespread interest has proven to be an efficient approach to measure the similarity between image pairs in quick and accurate manner by extracting complex and detailed features from satellite images. However, Image matching of VHR satellite images remains challenging due to limitations of DL models in which the results are depending on the quantity and quality of training dataset, as well as the difficulty of creating training dataset with VHR satellite images. Therefore, this study examines the feasibility of DL-based method in matching pair extraction which is the most time-consuming process during image registration. This paper also aims to analyze factors that affect the accuracy based on the configuration of training dataset, when developing training dataset from existing multi-sensor VHR image database with bias for DL-based image matching. For this purpose, the generated training dataset were composed of correct matching pairs and incorrect matching pairs by assigning true and false labels to image pairs extracted using a grid-based Scale Invariant Feature Transform (SIFT) algorithm for a total of 12 multi-temporal and multi-sensor VHR images. The Siamese convolutional neural network (SCNN), proposed for matching pair extraction on constructed training dataset, proceeds with model learning and measures similarities by passing two images in parallel to the two identical convolutional neural network structures. The results from this study confirm that data acquired from VHR satellite image database can be used as DL training dataset and indicate the potential to improve efficiency of the matching process by appropriate configuration of multi-sensor images. DL-based image matching techniques using multi-sensor VHR satellite images are expected to replace existing manual-based feature extraction methods based on its stable performance, thus further develop into an integrated DL-based image registration framework.

• Journal of Food Hygiene and Safety
• /
• v.38 no.2
• /
• pp.69-78
• /
• 2023

This study aimed to investigate the protective effect of enzymatically modified stevia (EMS) on C2C12 cell-based model of dexamethasone (DEX)-induced muscle atrophy to provide baseline data for utilizing EMS in functional health products. C2C12 cells with DEX-induced muscle atrophy were treated with EMS (10, 50, and 100 ㎍/mL) for 24 h. C2C12 cells were treated with EMS and DEX to test their effects on cell viability and myotube formation (myotube diameter and fusion index), and analyze the expression of muscle strengthening or degrading protein markers. Schisandra chinensis Extract, a common functional ingredient, was used as a positive control. EMS did not show any cytotoxic effect at all treatment concentrations. Moreover, it exerted protective effects on C2C12 cell-based model of DEX-induced muscle atrophy at all concentrations. In addition, the positive effect of EMS on myotube formation was confirmed based on the measurement and comparison of the fusion index and myotube diameter when compared with myotubes treated with DEX alone. EMS treatment reduced the expression of muscle cell degradation-related proteins Fbx32 and MuRF1, and increased the expression of muscle strengthening and synthesis related proteins SIRT1 and pAkt/Akt. Thus, EMS is a potential ingredient for developing functional health foods and should be further evaluated in preclinical models.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.68 no.3
• /
• pp.134-146
• /
• 2023

Phytophthora root rot (PRR) is a major soybean disease caused by an oomycete, Phytophthora sojae. PRR can be severe in poorly drained fields or wet soils. The disease management primarily relies on resistance genes called Rps (resistance to P. sojae). This study aimed to identify resistance loci associated with resistance to P. sojae isolate 40468 in Daepung × CheonAl recombinant inbred line (RIL) population. CheonAl is resistant to the isolate, while Daepung is generally susceptible. We genotyped the parents and RIL population via high-throughput single nucleotide polymorphism genotyping and constructed a set of genetic maps. The presence or absence of resistance to P. sojae was evaluated via hypocotyl inoculation technique, and phenotypic distribution fit to a ratio of 1:1 (R:S) (χ² = 0.57, p = 0.75), indicating single gene mediated inheritance. Single-marker association and the linkage analysis identified a highly significant genomic region of 55.9~56.4 megabase pairs on chromosome 18 that explained ~98% of phenotypic variance. Many previous studies have reported several Rps genes in this region, and also it contains nine genes that are annotated to code leucine-rich repeat or serine/threonine kinase within the approximate 500 kilobase pairs interval based on the reference genome database. CheonAl is the first domestic soybean genotype characterized for resistance against P. sojae isolate 40468. Therefore, CheonAl could be a valuable genetic source for breeding resistance to P. sojae.