• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.43-62
• /
• 2019

At one time, the anomaly detection sector dominated the method of determining whether there was an abnormality based on the statistics derived from specific data. This methodology was possible because the dimension of the data was simple in the past, so the classical statistical method could work effectively. However, as the characteristics of data have changed complexly in the era of big data, it has become more difficult to accurately analyze and predict the data that occurs throughout the industry in the conventional way. Therefore, SVM and Decision Tree based supervised learning algorithms were used. However, there is peculiarity that supervised learning based model can only accurately predict the test data, when the number of classes is equal to the number of normal classes and most of the data generated in the industry has unbalanced data class. Therefore, the predicted results are not always valid when supervised learning model is applied. In order to overcome these drawbacks, many studies now use the unsupervised learning-based model that is not influenced by class distribution, such as autoencoder or generative adversarial networks. In this paper, we propose a method to detect anomalies using generative adversarial networks. AnoGAN, introduced in the study of Thomas et al (2017), is a classification model that performs abnormal detection of medical images. It was composed of a Convolution Neural Net and was used in the field of detection. On the other hand, sequencing data abnormality detection using generative adversarial network is a lack of research papers compared to image data. Of course, in Li et al (2018), a study by Li et al (LSTM), a type of recurrent neural network, has proposed a model to classify the abnormities of numerical sequence data, but it has not been used for categorical sequence data, as well as feature matching method applied by salans et al.(2016). So it suggests that there are a number of studies to be tried on in the ideal classification of sequence data through a generative adversarial Network. In order to learn the sequence data, the structure of the generative adversarial networks is composed of LSTM, and the 2 stacked-LSTM of the generator is composed of 32-dim hidden unit layers and 64-dim hidden unit layers. The LSTM of the discriminator consists of 64-dim hidden unit layer were used. In the process of deriving abnormal scores from existing paper of Anomaly Detection for Sequence data, entropy values of probability of actual data are used in the process of deriving abnormal scores. but in this paper, as mentioned earlier, abnormal scores have been derived by using feature matching techniques. In addition, the process of optimizing latent variables was designed with LSTM to improve model performance. The modified form of generative adversarial model was more accurate in all experiments than the autoencoder in terms of precision and was approximately 7% higher in accuracy. In terms of Robustness, Generative adversarial networks also performed better than autoencoder. Because generative adversarial networks can learn data distribution from real categorical sequence data, Unaffected by a single normal data. But autoencoder is not. Result of Robustness test showed that he accuracy of the autocoder was 92%, the accuracy of the hostile neural network was 96%, and in terms of sensitivity, the autocoder was 40% and the hostile neural network was 51%. In this paper, experiments have also been conducted to show how much performance changes due to differences in the optimization structure of potential variables. As a result, the level of 1% was improved in terms of sensitivity. These results suggest that it presented a new perspective on optimizing latent variable that were relatively insignificant.

• Journal of agriculture & life science
• /
• v.53 no.5
• /
• pp.137-143
• /
• 2019

To manufacturing of solid fuel by reuse of the wastes, the drying unit which have 500 kg/hr of drying capacity was developed and experimentally evaluate the performance. The spinach grown in Nam-hae island were used for the experiments and investigated of the heated-air drying characteristics as the inlet amount of raw materials, raw material stirring status, conveying type and drying time. The drying air heated by the energy derived from the steam which is supplied from the incineration plant. The moisture contents of raw materials were measured 85.65%. The inlet flow rate of drying air made a difference as the depth of the raw materials loaded on the drying unit and temperature has showed 108~144℃. The drying speed of the mixed drying more than doubled as that of non mixed drying under the same drying type, inlet amount, drying time and drying air temperature. In each experiment, the drying capacity have showed over 500 kg/hr. A drying efficiency of the ratio of drying consumption energy to input energy was 33.46%, lower than the average of 57.76% for the 157 conventional dryers. Because developed dryer must have a drying time of less than one hour, it is considered that the dry efficiency has been reduced due to the loss of wind volume during drying. If waste heat from incineration plant is used as a direct heat source, the dry air temperature is expected to be at least 160℃, greatly improving the drying capacity.

• Clean Technology
• /
• v.29 no.1
• /
• pp.10-21
• /
• 2023

Hydrogels that are made of natural polymer-based double networks have excellent biocompatibility, low cytotoxicity, and high water content, assuring that the material has the properties required for a variety of biomedical applications. However, hydrogels also have limitations due to their relatively weak mechanical properties. In this study, hydrogels based on an alginate di-aldehyde (ADA) and gelatin (Gel) double network that is reinforced with additional hydrogen bonds formed between the hydroxyl (-OH) groups of the hyperbranched polymer (HPG) and the functional groups present inside of the hydrogels were successfully synthesized. The enhanced mechanical properties of these synthesized hydrogels were evaluated by varying the amount of HPG added during the hydrogel synthesis from 0 to 25%. In addition, magnetite nanoparticles (Fe₃O₄ NPs) were synthesized within the hydrogels and the structures and the magnetic properties of the hydrogels were also characterized. The hydrogels that contained 15% HPG and Fe₃O₄ NPs exhibited superparamagnetic behaviors with a saturation magnetization value of 3.8 emu g^-1. These particular hydrogels also had strengthened mechanical properties with a maximum compressive stress of 1.1 MPa at a strain of 67.4%. Magnetic hydrogels made with natural polymer-based double networks provide improved mechanical properties and have a significant potential for drug delivery and biomaterial application.

• Journal of IKEEE
• /
• v.27 no.1
• /
• pp.116-121
• /
• 2023

In this paper, we develop a deep learning structure for a complex microbial incubator that applies deep learning prediction result information. The proposed complex microbial incubator consists of pre-processing of complex microbial data, conversion of complex microbial data structure, design of deep learning network, learning of the designed deep learning network, and GUI development applied to the prototype. In the complex microbial data preprocessing, one-hot encoding is performed on the amount of molasses, nutrients, plant extract, salt, etc. required for microbial culture, and the maximum-minimum normalization method for the pH concentration measured as a result of the culture and the number of microbial cells to preprocess the data. In the complex microbial data structure conversion, the preprocessed data is converted into a graph structure by connecting the water temperature and the number of microbial cells, and then expressed as an adjacency matrix and attribute information to be used as input data for a deep learning network. In deep learning network design, complex microbial data is learned by designing a graph convolutional network specialized for graph structures. The designed deep learning network uses a cosine loss function to proceed with learning in the direction of minimizing the error that occurs during learning. GUI development applied to the prototype shows the target pH concentration (3.8 or less) and the number of cells (10⁸ or more) of complex microorganisms in an order suitable for culturing according to the water temperature selected by the user. In order to evaluate the performance of the proposed microbial incubator, the results of experiments conducted by authorized testing institutes showed that the average pH was 3.7 and the number of cells of complex microorganisms was 1.7 × 10⁸. Therefore, the effectiveness of the deep learning structure for the complex microbial incubator applying the deep learning prediction result information proposed in this paper was proven.

• Journal of Environmental Impact Assessment
• /
• v.32 no.6
• /
• pp.419-430
• /
• 2023

In response to the periodic occurrence of cyanobacterial blooms in Korean freshwaters, various types of cyanobacteria removal technologies are being developed and implemented. Due to the differing principles behind these technologies, it is difficult to compare and evaluate their removal efficiencies. In this study, a standardized method for evaluating cyanobacteria removal efficiency was proposed by utilizing the results of removal operations using a mobile cyanobacteria removal device in the Seohwacheon area of Daechung Reservoir. During removal operations, the decrease in chlorophyll-a (chl-a) concentration (ΔChl-a) in the working area was calculated based on the amount of collected sludge, the efficiency rate, and the concentration of chl-a. Additionally, the required working days (WD) to reduce the chl-a concentration to 1 mg/m³ in the target area was calculated based on the area of the target zone, the maximum daily working area, and the efficiency rate. A method for calculating the cyanobacteria removal capacity was proposed based on the reduction rate of chl-a concentration in the water before and after the operation, the treatment capacity of the removal technology, and the water volume of the target area. The cyanobacteria removal capacity of the mobile cyanobacteria removal device used in this study was 6.64%/day (targeting the Seohwacheon area of Daechung Reservoir, approximately 500,000 m²), which was higher compared to other physical or physicochemical cyanobacteria removal technologies (0.02~4.72%/day). Utilizing the evaluation method of cyanobacteria removal efficiency presented in this study, it will be possible to compare and evaluate the cyanobacteria removal technologies currently being applied in Korea. This method could also be used to assess the performance and efficiency of physical or physicochemical combined cyanobacteria removal techniques in the "Guidelines for the Installation and Operation of Algae Removal Facilities and the Use of Algae Removal Agents" operated by the National Institute of Environmental Research.

• Information Systems Review
• /
• v.16 no.3
• /
• pp.161-177
• /
• 2014

Corporate credit rating assessment consists of complicated processes in which various factors describing a company are taken into consideration. Such assessment is known to be very expensive since domain experts should be employed to assess the ratings. As a result, the data-driven corporate credit rating prediction using statistical and artificial intelligence (AI) techniques has received considerable attention from researchers and practitioners. In particular, statistical methods such as multiple discriminant analysis (MDA) and multinomial logistic regression analysis (MLOGIT), and AI methods including case-based reasoning (CBR), artificial neural network (ANN), and multiclass support vector machine (MSVM) have been applied to corporate credit rating.2) Among them, MSVM has recently become popular because of its robustness and high prediction accuracy. In this study, we propose a novel optimized MSVM model, and appy it to corporate credit rating prediction in order to enhance the accuracy. Our model, named 'GAMSVM (Genetic Algorithm-optimized Multiclass Support Vector Machine),' is designed to simultaneously optimize the kernel parameters and the feature subset selection. Prior studies like Lorena and de Carvalho (2008), and Chatterjee (2013) show that proper kernel parameters may improve the performance of MSVMs. Also, the results from the studies such as Shieh and Yang (2008) and Chatterjee (2013) imply that appropriate feature selection may lead to higher prediction accuracy. Based on these prior studies, we propose to apply GAMSVM to corporate credit rating prediction. As a tool for optimizing the kernel parameters and the feature subset selection, we suggest genetic algorithm (GA). GA is known as an efficient and effective search method that attempts to simulate the biological evolution phenomenon. By applying genetic operations such as selection, crossover, and mutation, it is designed to gradually improve the search results. Especially, mutation operator prevents GA from falling into the local optima, thus we can find the globally optimal or near-optimal solution using it. GA has popularly been applied to search optimal parameters or feature subset selections of AI techniques including MSVM. With these reasons, we also adopt GA as an optimization tool. To empirically validate the usefulness of GAMSVM, we applied it to a real-world case of credit rating in Korea. Our application is in bond rating, which is the most frequently studied area of credit rating for specific debt issues or other financial obligations. The experimental dataset was collected from a large credit rating company in South Korea. It contained 39 financial ratios of 1,295 companies in the manufacturing industry, and their credit ratings. Using various statistical methods including the one-way ANOVA and the stepwise MDA, we selected 14 financial ratios as the candidate independent variables. The dependent variable, i.e. credit rating, was labeled as four classes: 1(A1); 2(A2); 3(A3); 4(B and C). 80 percent of total data for each class was used for training, and remaining 20 percent was used for validation. And, to overcome small sample size, we applied five-fold cross validation to our dataset. In order to examine the competitiveness of the proposed model, we also experimented several comparative models including MDA, MLOGIT, CBR, ANN and MSVM. In case of MSVM, we adopted One-Against-One (OAO) and DAGSVM (Directed Acyclic Graph SVM) approaches because they are known to be the most accurate approaches among various MSVM approaches. GAMSVM was implemented using LIBSVM-an open-source software, and Evolver 5.5-a commercial software enables GA. Other comparative models were experimented using various statistical and AI packages such as SPSS for Windows, Neuroshell, and Microsoft Excel VBA (Visual Basic for Applications). Experimental results showed that the proposed model-GAMSVM-outperformed all the competitive models. In addition, the model was found to use less independent variables, but to show higher accuracy. In our experiments, five variables such as X7 (total debt), X9 (sales per employee), X13 (years after founded), X15 (accumulated earning to total asset), and X39 (the index related to the cash flows from operating activity) were found to be the most important factors in predicting the corporate credit ratings. However, the values of the finally selected kernel parameters were found to be almost same among the data subsets. To examine whether the predictive performance of GAMSVM was significantly greater than those of other models, we used the McNemar test. As a result, we found that GAMSVM was better than MDA, MLOGIT, CBR, and ANN at the 1% significance level, and better than OAO and DAGSVM at the 5% significance level.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.9 no.1
• /
• pp.1-18
• /
• 1973

For regulating the depth of midwater trawl nets towed at the optimum constant speed, the changes in the shape of warps caused by adding a weight on an arbitrary point of the warp of catenary shape is studied. The shape of a warp may be approximated by a catenary. The resultant inferences under this assumption were experimented. Accordingly feasibilities for the application of the result of this study to the midwater trawl nets were also discussed. A series of experiments for basic midwater trawl gear models in water tank and a couple of experiments of a commercial scale gears at sea which involve the properly designed depth control devices having a variable attitude horizontal wing were carried out. The results are summarized as follows: 1. According to the dimension analysis the depth y of a midwater trawl net is introduced by $$y=kLf(\frac{W_r}{R_r},\;\frac{W_o}{R_o},\;\frac{W_n}{R_n})$$) where k is a constant, L the warp length, f the function, and $W_r,\;W_o$ and $W_n$ the apparent weights of warp, otter board and the net, respectively, 2. When a boat is towing a body of apparent weight $W_n$ and its drag $D_n$ by means of a warp whose length L and apparent weight $W_r$ per unit length, the depth y of the body is given by the following equation, provided that the shape of a warp is a catenary and drag of the warp is neglected in comparison with the drag of the body: $$y=\frac{1}{W_r}\{\sqrt{{D_n^2}+{(W_n+W_rL)^2}}-\sqrt{{D_n^2+W_n}^2\}$$ 3. The changes ${\Delta}y$ of the depth of the midwater trawl net caused by changing the warp length or adding a weight ${\Delta}W_n$_n to the net, are given by the following equations: $${\Delta}y{\approx}\frac{W_n+W_{r}L}{\sqrt{D_n^2+(W_n+W_{r}L)^2}}{\Delta}L$$ $${\Delta}y{\approx}\frac{1}{W_r}\{\frac{W_n+W_rL}{\sqrt{D_n^2+(W_n+W_{r}L)^2}}-{\frac{W_n}{\sqrt{D_n^2+W_n^2}}\}{\Delta}W_n$$ 4. A change ${\Delta}y$ of the depth of the midwater trawl net by adding a weight $W_s$ to an arbitrary point of the warp takes an equation of the form $${\Delta}y=\frac{1}{W_r}\{(T_{ur}'-T_{ur})-T_u'-T_u)\}$$ Where $$T_{ur}^l=\sqrt{T_u^2+(W_s+W_{r}L)^2+2T_u(W_s+W_{r}L)sin{\theta}_u$$ $$T_{ur}=\sqrt{T_u^2+(W_{r}L)^2+2T_uW_{r}L\;sin{\theta}_u$$ $$T_{u}^l=\sqrt{T_u^2+W_s^2+2T_uW_{s}\;sin{\theta}_u$$ and $T_u$ represents the tension at the point on the warp, ${\theta}_u$ the angle between the direction of $T_u$ and horizontal axis, $T_u^2$ the tension at that point when a weights $W_s$ adds to the point where $T_u$ is acted on. 5. If otter boards were constructed lighter and adequate weights were added at their bottom to stabilize them, even they were the same shapes as those of bottom trawls, they were definitely applicable to the midwater trawl gears as the result of the experiments. 6. As the results of water tank tests the relationship between net height of H cm velocity of v m/sec, and that between hydrodynamic resistance of R kg and the velocity of a model net as shown in figure 6 are respectively given by $$H=8+\frac{10}{0.4+v}$$ $$R=3+9v^2$$ 7. It was found that the cross-wing type depth control devices were more stable in operation than that of the H-wing type as the results of the experiments at sea. 8. The hydrodynamic resistance of the net gear in midwater trawling is so large, and regarded as nearly the drag, that sweeping depth of the gear was very stable in spite of types of the depth control devices. 9. An area of the horizontal wing of the H-wing type depth control device was $1.2{\times}2.4m^2$. A midwater trawl net of 2 ton hydrodynamic resistance was connected to the devices and towed with the velocity of 2.3 kts. Under these conditions the depth change of about 20m of the trawl net was obtained by controlling an angle or attack of $30^{\circ}$.