• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.1
• /
• pp.114-123
• /
• 2018

This paper concerns a design and learning method of softmax function neural networks based on K-means clustering. The partial discharge data Information is preliminarily processed through simulation using an Epoxy Mica Coupling sensor and an internal Phase Resolved Partial Discharge Analysis algorithm. The obtained information is processed according to the characteristics of the pattern using a Motor Insulation Monitoring System program. At this time, the processed data are total 4 types that void discharge, corona discharge, surface discharge and slot discharge. The partial discharge data with high dimensional input variables are secondarily processed by principal component analysis method and reduced with keeping the characteristics of pattern as low dimensional input variables. And therefore, the pattern classifier processing speed exhibits improved effects. In addition, in the process of extracting the partial discharge data through the MIMS program, the magnitude of amplitude is divided into the maximum value and the average value, and two pattern characteristics are set and compared and analyzed. In the first half of the proposed partial discharge pattern classifier, the input and hidden layers are classified by using the K-means clustering method and the output of the hidden layer is obtained. In the latter part, the cross entropy error function is used for parameter learning between the hidden layer and the output layer. The final output layer is output as a normalized probability value between 0 and 1 using the softmax function. The advantage of using the softmax function is that it allows access and application of multiple class problems and stochastic interpretation. First of all, there is an advantage that one output value affects the remaining output value and its accompanying learning is accelerated. Also, to solve the overfitting problem, L2-normalization is applied. To prove the superiority of the proposed pattern classifier, we compare and analyze the classification rate with conventional radial basis function neural networks.

• Journal of Korean Library and Information Science Society
• /
• v.20
• /
• pp.351-381
• /
• 1993

The purpose of this study is to analyze the various processes and patterns to build or synthesize class numbers in the 800 class of the Dewey Decimal Classification, Edition 20(1989). The results of the analysis are as follows: 1. The 800(Literature and rhetoric) class in the DDC system is the main class added analytico-synthetic principle positively to an enumerative scheme. 2. The facets to be a n.0, pplied in literature are language literary form literary period ; kind, scope, or medium ; notation 08(collection) or 09(criticism) literary feature, subject, author, etc. 3. In the 800 class, there are the five tables of precedence for literary forms aspects ; specific kinds of persons ; literary, period in relation to the aspects for works treating more than one literary form subforms, aspects and literary periods in the works treating a specific literary form. 4. The basic number synthesis of literary works proceeds through the various facets in the following sequence, as far as necessary for the item : base no. + literary form + literary time or period + kind, scope, or medium + notation 08 or 09 + subform + additional notation from T3C and other tables. 5. In view of the multiplicity of facets, their synthesis formulas take the following order : (1) Works about the literature : base no.(schedule) + language(T6) or form(T3B) (2) Works by or about individual author : base no.(schedule) + form (T3A) + period(schedule) + subform(T3A) (3) Works by or about more than one author, not restricted by language facet : base no.(schedule) + period(T1) ; base no.(schedule) + kind, scope, medium(T3B), or feature(T3C), or person(T5). (4) Works by or about more than one author, restricted by language facet : base no.(schedule) + form (T3B) + period(schedule) + subform(T3B) + notation 08 or 09(T3B) ; base no.(schedule) + notation 08 or 09(T3B) + 9(T3C) + area notation(T2) : base no.(schedule) + form (T3B) + notation 008 or 009(T3B) : base no.(schedule) + form (T3B) + kind, scope, medium(T3B) + notation 08 or 09(T3B) + period(schedule). (5) Affiliated literatures for which period numbers are not us base no.(schedule) + form (T3A or T3B), or notation 08 or 09(T3B) : base no.(schedule) + kind, scope, medium(T3B), feature(T3C), or person(T5) 6. The problems in the number building of the 800 class are the complexity and difficulty of number synthesis, the intrinsic weakness of from distinction and the inconvenience of retrieval inherent in the form class. In order to solve these problems, therefore, the citation orders and methods of DDC should be improved and synthesis patterns simplified from the point of view of its applicability and its usefulness in the "literature class".

• The Journal of Korean Academy of Prosthodontics
• /
• v.36 no.2
• /
• pp.245-257
• /
• 1998

This study was performed to measure the changes of the mandibular movement and the masticatory muscular activities - anterior temporal and masseter muscle of both side - reflected by intentional increase of anterior guidance angie. For this study, 5 volunteers (3 males and 2 females with average age of 24.0) were selected. Each volunteer had Angle's classification I and did not have any missing tooth except third molar and any extensive restorations. Metallic guide plate was made at volunteer's working model fabricated by improved dental stone and cemented to the palatal surface of maxillary central incisor using resin cement(Panavia $21^{(R)}$) and then adjusted not to give any occlusal interferences at intercuspal position. The activity of masticatory muscles and the changes of mandibular movement were recorded by EMG and Sirognathograph in Biopak analysing system(Bioresearch Inc., Milwaukee, Wisconsin, USA). Measurement was done at before experiment, immediatley after placement, 1 week after placement, immediately after removal, and 1 week after removal. The results were as follows: 1. Moderate phonetic disturbance and mild headache were occured to 3 volunteers for 2 days after setting and 1 volunteer had positive reaction to percussion and slight midline diastema. But all of these clinical signs were diappeared 1 week after removal and the other volunteer did not have any special clinical sign. 2. In the EMG of the mandibular rest position, the mean value of anterior tempotal muscle was increased immediately after placement(p<0.01) and then decreased 1 week after placement(p<0.05) and increased 1 week after removal(p<0.05) but not recovered as before experiment. The mean value of masseter muscle was decreased during the experiment period. 3. In the EMG during mandibular protrusive movement, all muscular activity was decreased during the experiment period. Reduced activity was not recovered 1 week after removal(p<0.03). 4. During the habitual opening, anteroposterior movement of mandible was decreased immediately after placement(p<0.05) and then increased 1 week after placement but not statistically significant(p>0.1). Vertical movement was not shown significant difference during the experiment period(p>0.1). Lateral movement was decreased immediately after placement(p<0.05) and then increased 1 week after placement but not recovered as before experiment. The opening and closing velocity of mandible was shown minor changes but not statistically significant. 5. During the habitual opening, anteroposterior movement of mandible was decreased 1 week after placement(p<0.05) and then increased immediately after removal and recovered 1 week after removal as before experiment. Vertical movement was not shown significant changes. Lateral displacement of mandible was increased continuously and recovered 1 week after removal. Opening velocity was temporarily increased immediately after removal but recovered and closing velocity was not shown significant changes. 6. During the right side chewing, anteroposterior movement of mandible was increased immediately after removal but recovered and vertical movement was not shown statistically significant results. Lateral displacement and velocity of mandible were not shown significant results. 7. During the left side chewing, the changes of mandibular movement pattern were not shown statistically significant results.

• The Journal of the Acoustical Society of Korea
• /
• v.32 no.3
• /
• pp.252-261
• /
• 2013

The common speech recognition system displays higher recognition performance in a quiet environment, while its performance declines sharply in a real environment where there are noises. To implement a speech recognizer that is robust in different speech settings, this study suggests the method of Parallel Model Combination adaptation using frequency-variant based on environment-awareness (FV-PMC), which uses variants in frequency; acquires the environmental data for speech recognition; applies it to upgrading the speech recognition model; and promotes its performance enhancement. This FV-PMC performs the speech recognition with the recognition model which is generated as followings: i) calculating the average frequency variant in advance among the readily-classified noise groups and setting it as a threshold value; ii) recalculating the frequency variant among noise groups when speech with unknown noises are input; iii) regarding the speech higher than the threshold value of the relevant group as the speech including the noise of its group; and iv) using the speech that includes this noise group. When noises were classified with the proposed FV-PMC, the average accuracy of classification was 56%, and the results from the speech recognition experiments showed the average recognition rate of Set A was 79.05%, the rate of Set B 79.43%m, and the rate of Set C 83.37% respectively. The grand mean of recognition rate was 80.62%, which demonstrates 5.69% more improved effects than the recognition rate of 74.93% of the existing Parallel Model Combination with a clear model, meaning that the proposed method is effective.

• The Korean Journal of Archival Studies
• /
• no.13
• /
• pp.205-231
• /
• 2006

My suggestions in this paper come out of the review of the records on the Korean-Japanese negotiations(1952-1965). Before January 2002, the enforcement of the public records law, we had a poor management system of the diplomatic records. For a long time the diplomatic records of Korean government has not been preserved and managed according to the international and professional standards. So many important records have been probably lost and unsuitably classified, preserved for the future use. By the coming of public records law this deplorable situation in the management of diplomatic records has been much improved. However the registration, classification, compilation, based on the principle of provenance were not so sufficiently realized. It is now very urgent to employ more archivists in the relevant governmental institutions and organizations, and to introduce the concept of record group for the management of diplomatic papers. Also at the preparatory work for the publication of the diplomatic papers it is strongly needed to make a room for the participation of the civil experts such as historians, archivists and political scientists. In the case of publication of the Korean-Japanese papers it is also necessary to take the relevant American and Japanese governmental records on Korean-Japanese negotiations and private records of the actors of the times into account. Moreover it must be also seriously considered to start a big project for the elaborate edition of the important records of the foreign policy of the nation.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.155-170
• /
• 2018

Predicting a company's financial bankruptcy is traditionally one of the most crucial forecasting problems in business analytics. In previous studies, prediction models have been proposed by applying or combining statistical and machine learning-based techniques. In this paper, we propose a novel intelligent prediction model based on the simulated annealing which is one of the well-known optimization techniques. The simulated annealing is known to have comparable optimization performance to the genetic algorithms. Nevertheless, since there has been little research on the prediction and classification of business decision-making problems using the simulated annealing, it is meaningful to confirm the usefulness of the proposed model in business analytics. In this study, we use the combined model of simulated annealing and machine learning to select the input features of the bankruptcy prediction model. Typical types of combining optimization and machine learning techniques are feature selection, feature weighting, and instance selection. This study proposes a combining model for feature selection, which has been studied the most. In order to confirm the superiority of the proposed model in this study, we apply the real-world financial data of the Korean companies and analyze the results. The results show that the predictive accuracy of the proposed model is better than that of the naïve model. Notably, the performance is significantly improved as compared with the traditional decision tree, random forests, artificial neural network, SVM, and logistic regression analysis.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.42 no.5
• /
• pp.681-688
• /
• 2022

According to a recent study, most construction companies in Korea do not manage risk effectively, and it is judged that the risk management system needs to be improved. In addition, most risk-related studies deal with risks from a macroscopic perspective, and there are few studies dealing with process risks at the project construction stage. Therefore, this study tried to suggest a risk response plan through analysis and classification of risk factors that may occur in retaining work among process risks. To this end, a workshop was held for risk experts to identify and analyze risks that may occur during the construction of retaining work for apartments. As a result of the study, it was expected that savings of KRW 4.97 billion would be possible in the 95 % confidence interval, and the maximum possible cost was reduced from KRW 15 billion to about KRW 10 billion. Based on the risk reduction ratio, it was found that risks that can be reduced without any special input cost, risks with large effects in response to risks, and risks with insignificant effects were found. Therefore, using the types and risk factors presented in this study as guides, it is expected that it will be helpful in successfully operating the project if an appropriate response strategy is prepared and systematically responded to the site conditions.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.200-208
• /
• 2022

In this study, an automatic information collection system and geographic information construction algorithm for the transportation disadvantaged using electric mobility aids are implemented using an object recognition model. Recognizes objects that the disabled person encounters while moving, and acquires coordinate information. It provides an improved route selection map compared to the existing geographic information for the disabled. Data collection consists of a total of four layers including the HW layer. It collects image information and location information, transmits them to the server, recognizes, and extracts data necessary for geographic information generation through the process of classification. A driving experiment is conducted in an actual barrier-free zone, and during this process, it is confirmed how efficiently the algorithm for collecting actual data and generating geographic information is generated.The geographic information processing performance was confirmed to be 70.92 EA/s in the first round, 70.69 EA/s in the second round, and 70.98 EA/s in the third round, with an average of 70.86 EA/s in three experiments, and it took about 4 seconds to be reflected in the actual geographic information. From the experimental results, it was confirmed that the walking weak using electric mobility aids can drive safely using new geographic information provided faster than now.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.217-242
• /
• 2022

With the launch of Artificial Intelligence(AI)-based intelligent products on the market, innovative changes are taking place not only in business but also in consumers' daily lives. Intelligent products have the potential to realize technology differentiation and increase market competitiveness through advanced functions of artificial intelligence. However, there is no new product development methodology that can sufficiently reflect the characteristics of artificial intelligence for the purpose of developing intelligent products with high market acceptance. This study proposes a KANO-QFD integrated model as a methodology for intelligent product development. As a specific example of the empirical analysis, the types of consumer requirements for hair loss prediction and treatment device were classified, and the relative importance and priority of engineering characteristics were derived to suggest the direction of intelligent medical product development. As a result of a survey of 130 consumers, accurate prediction of future hair loss progress, future hair loss and improved future after treatment realized and viewed on a smartphone, sophisticated design, and treatment using laser and LED combined light energy were realized as attractive quality factors among the KANO categories. As a result of the analysis based on House of Quality of QFD, learning data for hair loss diagnosis and prediction, micro camera resolution for scalp scan, hair loss type classification model, customized personal account management, and hair loss progress diagnosis model were derived. This study is significant in that it presented directions for the development of artificial intelligence-based intelligent medical product that were not previously preceded.

• Journal of Korean Society of Rural Planning
• /
• v.28 no.1
• /
• pp.17-26
• /
• 2022

It is necessary to prepare a stable production base in advance for a change in the global grain market, and it is required to prepare comprehensive countermeasures such as securing technical skills and cultivation technology. Therefore, Korea, which relies on imports of major grains other than rice, could be exposed to a food crisis at any time unless the self-sufficiency rate of grains is improved. In order to respond to this new food crisis, it is necessary to find ways to efficiently utilize rice fields to increase the domestic grain self-sufficiency rate. From this point of view, interest and demand for the generalization of farmland that can be used as paddy fields and returned to paddy fields are increasing, and related research is also being continuously performed. In order to select a multipurpose farmland project site, this study extracted farmland containing 10% or more purchased and stockpiled farmland through spatial analysis (buffer, dissolve, intersect, etc.), and finally presented areas subject to multipurpose farmland projects. The target site for the multipurpose farmland project was finally selected by integrating data onto a point-by-point basis so that the current status of farmland purchased and stockpiled, Farm Manager Registration Records, and the Korean Soil Information System data (drainage classes, surface soil texture, field-suitability classification, etc.) can be used in combination. There are 175 areas where the multipurpose farmland is possible. Incheon 2, Gyeongbuk 40, Gangwon 2, Chungbuk 7, Chungnam 48, Jeonbuk 34, Jeonnam 19, Gyeongbuk 15, Gyeongnam 8. Chungcheongnam-do has the most target site for the multipurpose farmland project, and Gangwon-do is the least. It is expected to contribute to new commercialization and business expansion by deriving business areas by identifying the scale of the farmland multipurpose farmland project using Farm Manger Registration Records and spatial data.