• 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.17 no.1
• /
• pp.53-69
• /
• 2011

Vision and voice-based technologies are commonly utilized for human-robot interaction. But it is widely recognized that the performance of vision and voice-based interaction systems is deteriorated by a large margin in the real-world situations due to environmental and user variances. Human users need to be very cooperative to get reasonable performance, which significantly limits the usability of the vision and voice-based human-robot interaction technologies. As a result, touch screens are still the major medium of human-robot interaction for the real-world applications. To empower the usability of robots for various services, alternative interaction technologies should be developed to complement the problems of vision and voice-based technologies. In this paper, we propose the use of accelerometer-based gesture interface as one of the alternative technologies, because accelerometers are effective in detecting the movements of human body, while their performance is not limited by environmental contexts such as lighting conditions or camera's field-of-view. Moreover, accelerometers are widely available nowadays in many mobile devices. We tackle the problem of classifying acceleration signal patterns of 26 English alphabets, which is one of the essential repertoires for the realization of education services based on robots. Recognizing 26 English handwriting patterns based on accelerometers is a very difficult task to take over because of its large scale of pattern classes and the complexity of each pattern. The most difficult problem that has been undertaken which is similar to our problem was recognizing acceleration signal patterns of 10 handwritten digits. Most previous studies dealt with pattern sets of 8~10 simple and easily distinguishable gestures that are useful for controlling home appliances, computer applications, robots etc. Good features are essential for the success of pattern recognition. To promote the discriminative power upon complex English alphabet patterns, we extracted 'motion trajectories' out of input acceleration signal and used them as the main feature. Investigative experiments showed that classifiers based on trajectory performed 3%~5% better than those with raw features e.g. acceleration signal itself or statistical figures. To minimize the distortion of trajectories, we applied a simple but effective set of smoothing filters and band-pass filters. It is well known that acceleration patterns for the same gesture is very different among different performers. To tackle the problem, online incremental learning is applied for our system to make it adaptive to the users' distinctive motion properties. Our system is based on instance-based learning (IBL) where each training sample is memorized as a reference pattern. Brute-force incremental learning in IBL continuously accumulates reference patterns, which is a problem because it not only slows down the classification but also downgrades the recall performance. Regarding the latter phenomenon, we observed a tendency that as the number of reference patterns grows, some reference patterns contribute more to the false positive classification. Thus, we devised an algorithm for optimizing the reference pattern set based on the positive and negative contribution of each reference pattern. The algorithm is performed periodically to remove reference patterns that have a very low positive contribution or a high negative contribution. Experiments were performed on 6500 gesture patterns collected from 50 adults of 30~50 years old. Each alphabet was performed 5 times per participant using $Nintendo{(R)}$ $Wii^{TM}$ remote. Acceleration signal was sampled in 100hz on 3 axes. Mean recall rate for all the alphabets was 95.48%. Some alphabets recorded very low recall rate and exhibited very high pairwise confusion rate. Major confusion pairs are D(88%) and P(74%), I(81%) and U(75%), N(88%) and W(100%). Though W was recalled perfectly, it contributed much to the false positive classification of N. By comparison with major previous results from VTT (96% for 8 control gestures), CMU (97% for 10 control gestures) and Samsung Electronics(97% for 10 digits and a control gesture), we could find that the performance of our system is superior regarding the number of pattern classes and the complexity of patterns. Using our gesture interaction system, we conducted 2 case studies of robot-based edutainment services. The services were implemented on various robot platforms and mobile devices including $iPhone^{TM}$. The participating children exhibited improved concentration and active reaction on the service with our gesture interface. To prove the effectiveness of our gesture interface, a test was taken by the children after experiencing an English teaching service. The test result showed that those who played with the gesture interface-based robot content marked 10% better score than those with conventional teaching. We conclude that the accelerometer-based gesture interface is a promising technology for flourishing real-world robot-based services and content by complementing the limits of today's conventional interfaces e.g. touch screen, vision and voice.

• Journal of Intelligence and Information Systems
• /
• v.25 no.2
• /
• pp.25-38
• /
• 2019

Selecting high-quality information that meets the interests and needs of users among the overflowing contents is becoming more important as the generation continues. In the flood of information, efforts to reflect the intention of the user in the search result better are being tried, rather than recognizing the information request as a simple string. Also, large IT companies such as Google and Microsoft focus on developing knowledge-based technologies including search engines which provide users with satisfaction and convenience. Especially, the finance is one of the fields expected to have the usefulness and potential of text data analysis because it's constantly generating new information, and the earlier the information is, the more valuable it is. Automatic knowledge extraction can be effective in areas where information flow is vast, such as financial sector, and new information continues to emerge. However, there are several practical difficulties faced by automatic knowledge extraction. First, there are difficulties in making corpus from different fields with same algorithm, and it is difficult to extract good quality triple. Second, it becomes more difficult to produce labeled text data by people if the extent and scope of knowledge increases and patterns are constantly updated. Third, performance evaluation is difficult due to the characteristics of unsupervised learning. Finally, problem definition for automatic knowledge extraction is not easy because of ambiguous conceptual characteristics of knowledge. So, in order to overcome limits described above and improve the semantic performance of stock-related information searching, this study attempts to extract the knowledge entity by using neural tensor network and evaluate the performance of them. Different from other references, the purpose of this study is to extract knowledge entity which is related to individual stock items. Various but relatively simple data processing methods are applied in the presented model to solve the problems of previous researches and to enhance the effectiveness of the model. From these processes, this study has the following three significances. First, A practical and simple automatic knowledge extraction method that can be applied. Second, the possibility of performance evaluation is presented through simple problem definition. Finally, the expressiveness of the knowledge increased by generating input data on a sentence basis without complex morphological analysis. The results of the empirical analysis and objective performance evaluation method are also presented. The empirical study to confirm the usefulness of the presented model, experts' reports about individual 30 stocks which are top 30 items based on frequency of publication from May 30, 2017 to May 21, 2018 are used. the total number of reports are 5,600, and 3,074 reports, which accounts about 55% of the total, is designated as a training set, and other 45% of reports are designated as a testing set. Before constructing the model, all reports of a training set are classified by stocks, and their entities are extracted using named entity recognition tool which is the KKMA. for each stocks, top 100 entities based on appearance frequency are selected, and become vectorized using one-hot encoding. After that, by using neural tensor network, the same number of score functions as stocks are trained. Thus, if a new entity from a testing set appears, we can try to calculate the score by putting it into every single score function, and the stock of the function with the highest score is predicted as the related item with the entity. To evaluate presented models, we confirm prediction power and determining whether the score functions are well constructed by calculating hit ratio for all reports of testing set. As a result of the empirical study, the presented model shows 69.3% hit accuracy for testing set which consists of 2,526 reports. this hit ratio is meaningfully high despite of some constraints for conducting research. Looking at the prediction performance of the model for each stocks, only 3 stocks, which are LG ELECTRONICS, KiaMtr, and Mando, show extremely low performance than average. this result maybe due to the interference effect with other similar items and generation of new knowledge. In this paper, we propose a methodology to find out key entities or their combinations which are necessary to search related information in accordance with the user's investment intention. Graph data is generated by using only the named entity recognition tool and applied to the neural tensor network without learning corpus or word vectors for the field. From the empirical test, we confirm the effectiveness of the presented model as described above. However, there also exist some limits and things to complement. Representatively, the phenomenon that the model performance is especially bad for only some stocks shows the need for further researches. Finally, through the empirical study, we confirmed that the learning method presented in this study can be used for the purpose of matching the new text information semantically with the related stocks.

• Journal of Korea Water Resources Association
• /
• v.48 no.8
• /
• pp.613-623
• /
• 2015

To overcome the difficulties of discharge measurements during flood season, MWSCM(micowave water surface current meter) which measures river surface velocities without contacting water has been applied in field work since its development. The existing version of MWSCM is for floods so that its applicability is low due to the short periods of floods. Therefore the renovative redesign of MWSCM to increase the applicability was conducted so that it can be applied to the discharge measurements during normal flows as well as flood ones by extending the measurable range of velocity. A newly developed high-performance MWSCM for general use can measure the velocity range of 0.03-20.0 m/s from flood flows to normal flows, whereas MWSCM for floods can measure the velocity range of 0.5-10.0 m/s. The improvement of antenna isolation between transmitter and receiver to block the inflow of transmitted singals to receiver and the improvement of phase noise of oscillator are necessary for detecting low velocity with MWSCM technology. Separate type antenna of transmitting and receiving signals is developed for isolation enhancement and phase locked loop synthesizer as an oscillator is applied to high-performance MWSCM for general use. Microwave frequency of 24 GHz is applied to the new MWSCM rather than 10 GHz to make the new MWSCM small and light for convenient use of it at fields. Improvement requests on MWSCM for floods-stable velocity measurement, self test, low power consumtion, and waterproof and dampproof-from the users of it has been reflected on the development of the new version of MWSCM.

• Journal of Bio-Environment Control
• /
• v.26 no.4
• /
• pp.353-360
• /
• 2017

In this study, the design and performance test of the air to water heat pump capable of producing hot water for greenhouse heating by using the surplus solar heat inside the greenhouse and the air heat outside greenhouse as the selective heat source were conducted. The heat storage operations using the surplus solar heat and the outside air heat were designed to be switched according to the setting temperature of the greenhouse in consideration of the optimum temperature range of the crop. In the developed system, it was possible to automatically control the switching of heat storage operation, heating and ventilation by setting 12 reference temperatures on the control panel. In the selective heat storage operation with the surplus solar heat and outside air heat, the temperature of thermal storage tank was controlled variably from $35^{\circ}C$ to $52^{\circ}C$ according to the heat storage rate and heating load. The heat storage operation times using the surplus solar heat and outside air heat were 23.1% and 30.7% of the experimental time respectively and the heat pump pause time was 46.2%. COP(coefficient of performance) of the heat pump of the heat storage operation using the surplus solar heat and outside air heat were 3.83 and 2.77 respectively and was 3.24 for whole selective heat storage operation. For the comparative experiment, the heat storage operation using the outside air heat only was performed under the condition that the temperature of the thermal storage tank was controlled constantly from 50 to $52^{\circ}C$, and COP was analyzed to be 2.33. As a result, it was confirmed that the COP of the heat storage operation using the surplus solar heat and outside air heat as selective heat source and the variable temperature control of the thermal storage tank was 39% higher than that of the general heat storage operation using the outside air heat only and the constant temperature control of the thermal storage tank.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.39 no.4
• /
• pp.347-357
• /
• 2003

This paper describes on the suppression of sea clutter on marine radar display using a cell-averaging CFAR(constant false alarm rate) technique, and on the analysis of radar echo signal data in relation to the estimation of ARPA functions and the detection of the shadow effect in clutter returns. The echo signal was measured using a X -band radar, that is located on the Pukyong National University, with a horizontal beamwidth of $$3.9^{\circ}$$, a vertical beamwidth of $20^{\circ}$, pulsewidth of $0.8 {\mu}s$ and a transmitted peak power of 4 ㎾ The suppression performance of sea clutter was investigated for the probability of false alarm between $l0-^0.25;and; 10^-1.0$. Also the performance of cell averaging CFAR was compared with that of ideal fixed threshold. The motion vectors and trajectory of ships was extracted and the shadow effect in clutter returns was analyzed. The results obtained are summarized as follows；1. The ARPA plotting results and motion vectors for acquired targets extracted by analyzing the echo signal data were displayed on the PC based radar system and the continuous trajectory of ships was tracked in real time. 2. To suppress the sea clutter under noisy environment, a cell averaging CFAR processor having total CFAR window of 47 samples(20+20 reference cells, 3+3 guard cells and the cell under test) was designed. On a particular data set acquired at Suyong Man, Busan, Korea, when the probability of false alarm applied to the designed cell averaging CFAR processor was 10$^{-0}$.75/ the suppression performance of radar clutter was significantly improved. The results obtained suggest that the designed cell averaging CFAR processor was very effective in uniform clutter environments. 3. It is concluded that the cell averaging CF AR may be able to give a considerable improvement in suppression performance of uniform sea clutter compared to the ideal fixed threshold. 4. The effective height of target, that was estimated by analyzing the shadow effect in clutter returns for a number of range bins behind the target as seen from the radar antenna, was approximately 1.2 m and the information for this height can be used to extract the shape parameter of tracked target..

• 대한공업교육학회지
• /
• v.40 no.1
• /
• pp.142-162
• /
• 2015

The purpose of this study are to provide the basic data materials and implementations for successful performance of electric-work field representatives of South Korean firms by identifying their roles and competency and examining their educational need. For this research purposes, three phased analysis was followed on: (1) the roles of electric-work field representatives, (2) competency of electric-work field representatives and (3) educational need for their competency. This research method was to conduct a focus group interview for 10 expert field representatives along with survey. The collected data materials were processed by MS Excel and SPSS 21.0 for statistical analysis including average, standard deviation and other basic statistics; the gap in awareness of field representatives; and need values. For the needs analysis, the difference between significance of field representatives' competency and current status was examined by t test. And the awareness gap between competency importance and current status was identified based on the Borich equation. The Locus for Focus model was employed herein to identify the kinds of competency with high importance and high inconsistency to prioritize. As a result, this research has found as follows: first, the roles of field representatives were found to be in 13 different kinds of roles. Second, electric-work field representatives were found to need to have 16 different skills. Third, regarding the 16 abilities, the gap between current status and significance was analyzed herein. The results showed statistically significant differences in all cases. The Borich needs analysis found the first required ability was communication ability followed by power of execution, conflict management ability, analytical thinking and time management ability. Also, the results of Locus for Focus model analysis displayed that the first quadrant(HH) included 7 highly-demanded abilities of communication ability, analytical thinking, decision making ability, specialty, time management ability, power of execution and drive for work implementation. The top-priority group was found to have 5 items of communication ability, analytical thinking, time management ability, power of execution and drive for work implementation which were commonly seen in the Locus for Focus model outcomes. Based on these findings, this research could identify the roles and competency of electric-work field representatives and provide the basic data materials applicable to future personal management of electricity companies including recruitment, division of work, job description, evaluation, etc. Also this research offered guidelines on demanded abilities in the field and where to place priority. The kinds of abilities with high educational demand as found in this research must be considered in designing educational programs for the competency building of field representatives. This research is expected to provide useful information in developing such educational programs for field representatives.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.16 no.1
• /
• pp.3293-3301
• /
• 1974

An experimental work was conducted by using a laboratory-made model dryer to investigate the effect of the rate of natural forced-air on the drying rate of rough rice which was deposited in the deep-bed. The dryer consisted of 8 cylinderical containers with grain holding screen at their bottoms, each of which having 30cm in diameter and 15cm in height. The containers were sacked vertically with keeping them air-tight by using paper tape during dryer operation. Two separate layers of containers were operated in the same time to have two replications. The moisture contents of grains within each bins after predetermined period of dryer operation were determined indirectly by measuring the weight of the individual containers. The air-rates were maintained at 6 levels, or 5, 8, 10, 15, 18 and 20 millimenters of static head of water. The roomair conditions during dryer operation were maintained in the range of 10-l5$^{\circ}C$ in temperature and 40-60% in relative humidity. The results of the study are summarized as follows: 1. Drying characteristics of the grains in the bottom layers were approximately the same regardless of airdelivery rates, giving the average drying rate as about 0.35 percent per hour after 40-hour drying period, during which moisture content (w. b.) reduced from 24 percent to about 10 percent. 2. After about 40-hour drying period, the mean drying rates increased from 0.163 percent per hour to 0.263 percent per hour as air-flow rates increased from 5mm to 87.16mm of static head of water. In the same time, the moisture differences of grains between lower and upper layers varied from 12.7 percent at the air rate of 5mm of water head to 7.5 percent at the air-flow rate of 20mn of water head. Thus, the greater the air-flow rate was, the more overall improvement in drying performance was. Additionally, from the result of ineffectiveness of drying grain positioned at 70cm depth or above by the air rate of 5mm of static head of water it may be suggested in practical application that the height of grain deposit would be maintained adequately within the limits of air-rates that may be actually delivered. 3. Drying after layer-turning operation was continued for about 30 hours to test the effectiveness of reducing moisture differences in the thick layers. As a result of this layer-turning operation, moisture distribution through layers approached to narrow ranges, giving the moisture range as about 7 percent at air-flow rate of 5mm head of water, about 3 percent at 10mm head about 2 percent at 15mm head, and less than 1 percent at 20mm head. In addition, from the desirable results that drying rate was rapid in the lower layers and dully in the upper layers, layer-turning operation may be very effective in natural air drying with deep-layer grain deposit, especially when the forced air was kept in low rate. 4. Even though the high rate of air delivery is very desirable for deep-layer natural-air drying of rough rice, it can be happened that the required air delivery rate could not be attained because of limitation of power source available on farms. To give a guide line for the practical application, the power required to perform the drying with the specified air rate was analyzed for different sizes of drying bin and is given in Table (5). If a farmer selects a motor of which size is 1 or {{{{1 { 1} over {2 } }}}} H.P. and air-delivery rate which ranges from 8~10mm of head, the diameter of grain bin may be suggested to choose about 2.4m, also power tiller or other moderate size of prime motor may be recommended when the diameter of grain bin is about 5.0m or more for about 120cm grain deposit.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.5
• /
• pp.479-487
• /
• 2023

Recently, the destructive power of typhoons is continuously increasing owing to global warming. In a situation where the installation of floating wind turbines is increasing worldwide, concerns about the huge loss and collapse of floating offshore wind turbines owing to strong typhoons are deepening. A new type of disconnectable mooring system must be developed for the safe operation of floating offshore wind turbines. A new submersible mooring pulley considered in this study is devised to more easily attach or detach the floating of shore wind turbine with mooring lines compared with other disconnectable mooring apparatuses. To investigate the structural safety of the initial design of submersible mooring pulley that can be applied to an 8MW-class floating type offshore wind turbine, scale-down structural models were developed using a 3-D printer and structural tests were performed on the models. For the structural tests of the scale-down models, tensile specimens of acrylonitrile butadiene styrene material that was used in the 3-D printing were prepared, and the material properties were evaluated by conducting the tensile tests. The finite element analysis (FEA) of submersible mooring pulley was performed by applying the material properties obtained from the tensile tests and the same load and boundary conditions as in the scale-down model structural tests. Through the FEA, the structural weak parts on the submersible mooring pulley were reviewed. The structural model tests were conducted considering the main load conditions of submersible mooring pulley, and the FEA and test results were compared for the locations that exceeded the maximum tensile stress of the material. The results of the FEA and structural model tests indicated that the connection structure of the body and the wheel was weak in operating conditions and that of the body and the chain stopper was weak in mooring conditions. The results of this study enabled to experimentally verify the structural safety of the initial design of submersible mooring pulley. The study results can be usefully used to improve the structural strength of submersible mooring pulley in a detailed design stage.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.37 no.8
• /
• pp.781-788
• /
• 2013

Predicting the engine component temperature is a basic step to conduct structural safety evaluation in medium-speed diesel engine design. Recent trends such as increasing power density and performance necessitate more effective thermal management of the engine for achieving the desired durability and reliability. In addition, the local temperatures of several engine components must be maintained in the proper range to avoid problems such as low- or high-temperature corrosion. Therefore, it is very important to predict the temperature distribution of each engine part accurately in the design stage. In this study, the temperature of an engine component is calculated by using steady-state conjugate heat transfer analysis. A proper approach to determine the thermal load distribution on the thermal boundary area is suggested by using 1D engine system analysis, 3D transient CFD results, and previous experimental data from another developed engine model. A Hyundai HiMSEN engine having 250-mm bore size was chosen to validate the analysis procedure. The predicted results showed a reasonable agreement with experimental results.