• Imaging Science in Dentistry
• /
• v.47 no.3
• /
• pp.149-155
• /
• 2017

Purpose: The classification of impacted maxillary third molars(IMTMs) facilitates interdisciplinary communication and helps estimate the degree of surgical difficulty. Thus, this study aimed to develop a 3-dimensional classification of the position and depth of impaction of IMTMs and to estimate their prevalence with respect to gender and age. Materials and Methods: This cross-sectional retrospective study analyzed images in sagittal and coronal cone-beam computed tomography (CBCT) sections of 300 maxillary third molars. The proposed classification was based on 3 criteria: buccolingual position (buccal, lingual, or central), mesial-distal position (mesioangular, vertical, or distoangular), and depth of impaction (low, medium, or high). CBCT images of IMTMs were classified, and the associations of the classifications with gender and age were examined using analysis of variance with the Scheffe post-hoc test. To determine the associations among the 3 classifications, the chi-square test was used (P<.05). Results: No significant association of the classifications with gender was observed. Age showed a significant relationship with depth of impaction (P=.0001) and mesial-distal position (P=.005). The most common positions were buccal(n=222), vertical(n=184), and low (n=124). Significant associations among the 3 tested classifications were observed. Conclusion: CBCT enabled the evaluation of IMTMs in a 3-dimensional format, and we developed a proposal for a new classification of the position and depth of impaction of IMTMs.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.2
• /
• pp.830-839
• /
• 2017

Support vector machine (SVM) is introduced as an effective fault diagnosis technique based on dissolved gases analysis (DGA) for oil-immersed transformers with maximum generalization ability; however, the applicability of the SVM is highly affected due to the difficulty of selecting the SVM parameters appropriately. Therefore, a novel approach combing SVM with improved imperialist competitive algorithm (IICA) for fault diagnosis of oil-immersed transformers was proposed in the paper. The improved ICA, which is proved to be an effective optimization approach, is employed to optimize the parameters of SVM. Cross validation and normalizations were applied in the training processes of SVM and the trained SVM model with the optimized parameters was established for fault diagnosis of oil-immersed transformers. Three classification benchmark sets were studied based on particle swarm optimization SVM (PSOSVM) and IICASVM with four multiple classification schemes to select the best scheme for transformer fault diagnosis. The results show that the proposed model can obtain higher diagnosis accuracy than other methods. The comparisons confirm that the proposed model is an effective approach for classification problems.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.163-170
• /
• 2021

As a big data is being used in various industries, big data market is expanding from hardware to infrastructure software to service software. Especially it is expanding into a huge platform market that provides applications for holistic and intuitive visualizations such as big data meaning interpretation understandability, and analysis results. Demand for big data extraction and analysis using social media such as SNS is very active not only for companies but also for individuals. However despite such high demand for the collection and analysis of social media data for user trend analysis and marketing, there is a lack of research to address the difficulty of dynamic interlocking and the complexity of building and operating software platforms due to the heterogeneity of various social media service interfaces. In this paper, we propose a method for developing a framework to operate the process from collection to extraction and classification of social media data. The proposed framework solves the problem of heterogeneous social media data collection channels through adapter patterns, and improves the accuracy of social topic extraction and classification through semantic association-based extraction techniques and topic association-based classification techniques.

• The Korean Journal of Archival Studies
• /
• no.3
• /
• pp.65-89
• /
• 2001

We, at The Korea Democracy Movement Archives opened temporarily at SungKongHoe University(SKHU), have currently collected about 100,000 recorded materials of democratization movement related with labor, farmer, civilian, human rights, peace, unification, young people, student and women's movements by investigating, collecting and receiving donations from civil organizations and individuals, and about 70,000 data out of this 100,000 data were converted into computer files. The Korea Democracy Movement Archives(temp) at SKHU has a significance in that it is the first archive opened by an organization. Furthermore, the opening of this Archive means the expansion of awareness on recording culture and accumulation of the achievements of the democratization movement in Korea. However, many obstacles still remain in the establishment of this Archive in a full-scale. This article examined many theoretical and realistic obstacles posed to the archivists, who are the professionals responsible for record management, in process of establishing the Archive, and the role and future perspectives of the archivists at The Korea Democracy Movement Archives(temp). The first obstacles in the process of organizing and separating the recorded materials at the Archive is a difficulty in the description of classifying the different movement organizations. The second obstacle is a difficulty in specifically applying the international standard, ISAD(G), of record description in the process of establishing the description items. Through many trials and errors, we need to try to confirm the description befitting. The Korea Democracy Movement Archives through continuous adjustment and complementary measures. The third obstacle is a difficulty in estimating the range and physical and quantitative amount of the recorded materials since the collection of recorded materials is complete. Thus, the answers to these problems lie in continuous efforts to establish a creative classification system befitting the democratization movement in Korea in the process of many trials and errors and endeavor. The evaluation classification done by archivists is a creative act forming record heritage, and archivists need to form record heritage reflecting the evaluation system of a certain period. Moreover, they transmit the shape of the current era in a maximum scale to the future by using the minimum amount of records. An archivist is responsible for two tasks, i.e., preserving a record and making other people to utilize the record by working with record. However, We, at The Korea Democracy Movement Archives(temp), have an additional task of contributing to the democratization movement in korea that has not ended by collecting, preserving and making people to utilize the fragments of memory in the recent history of Korea by establishing the Archives.

• KIISE Transactions on Computing Practices
• /
• v.22 no.9
• /
• pp.455-460
• /
• 2016

Recently, with more severe types felonies such as robbery and sexual violence, the importance of crime prediction and prevention is emphasized. For accurate and prompt crime prediction and prevention, both a classification model of crime with high accuracy based on past criminal records and well-designed system interface are required. However previous studies on the analysis of crime factors have limitations in terms of accuracy due to the difficulty of data preprocessing. In addition, existing crime monitoring systems merely offer a vast amount of crime analysis results, thereby they fail to provide users with functions for more effective monitoring. In this paper, we propose a classification model for types of crime based on random-forest algorithms and system design factors for real-time crime prediction. From our experiments, we proved that our proposed classification model is superior to others that only use criminal records in terms of accuracy. Through the analysis of existing crime monitoring systems, we also designed and developed a system for real-time crime monitoring.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.43-54
• /
• 2024

Recently, there has been a lot of research on utilizing Language Models, and Large Language Models have achieved innovative results in various tasks. However, the practical application faces limitations due to the constrained resources and costs required to utilize Large Language Models. Consequently, there has been recent attention towards methods to effectively utilize models within given resources. Curriculum Learning, a methodology that categorizes training data according to difficulty and learns sequentially, has been attracting attention, but it has the limitation that the method of measuring difficulty is complex or not universal. Therefore, in this study, we propose a methodology based on data heterogeneity-based Curriculum Learning that measures the difficulty of data using reliable prior information and facilitates easy utilization across various tasks. To evaluate the performance of the proposed methodology, experiments were conducted using 5,000 specialized documents in the field of information communication technology and 4,917 documents in the field of healthcare. The results confirm that the proposed methodology outperforms traditional fine-tuning in terms of classification accuracy in both LoRA fine-tuning and full fine-tuning.

• Korean Journal of Construction Engineering and Management
• /
• v.25 no.1
• /
• pp.32-41
• /
• 2024

In the construction industry, not only safety accidents, but also various complex risks such as construction delays, cost increases, and environmental pollution occur, and management technologies are needed to solve them. Among them, process risk management, which directly affects the project, lacks related information compared to its importance. This study tried to develop a MATM tunnel process risk classification system to solve the difficulty of risk information retrieval due to the use of different classification systems for each project. Risk collection used existing literature review and experience mining techniques, and DB construction utilized the concept of natural language processing. For the structure of the classification system, the existing WBS structure was adopted in consideration of compatibility of data, and an RBS linked to the work species of the WBS was established. As a result of the research, a risk classification system was completed that easily identifies risks by work type and intuitively reveals risk characteristics and risk factors linked to risks. As a result of verifying the usability of the established classification system, it was found that the classification system was effective as risks and risk factors for each work type were easily identified by user input of keywords. Through this study, it is expected to contribute to preventing an increase in cost and construction period by identifying risks according to work types in advance when planning and designing NATM tunnels and establishing countermeasures suitable for those factors.

• The Korean Journal of Archival Studies
• /
• no.1
• /
• pp.137-188
• /
• 2000

Korea was highly successful in maintaining a strong tradition of record preservation. However, Japanese imperialism and the Korean war over this century has seen a huge quantity of these once plentiful records either destroyed or relocated overseas. Many of Korea records remain in the custody of foreign archival institutions. Additionally, many records of foreign states concerned with korean affairs, for example, records of the Department of state, in the U.S.A., are of great relevance to Korean modern history. These records which exist in foreign countries are a necessary and invaluable resource in the advancement of Korea studies. After the 1980s, as research interest in Korean modern history increased gradually, much historical material related to Korea was extensively introduced, removed, collected from the U.S.A., Japan, Russia and China etc. by the efforts of pioneering scholars and institutions. Several attempts at collating and publishing this material have been made. Despite this encouraging result, individual scholars and institutions's efforts face problems in relation to removal of records. Firstly, it is repeated nonproductively to remove and collect the same records, for we didn't establish a comprehensive information system, through which the condition of foreign record removal and management is able to be administrated. Secondly, there is no system for arrangement, preservation, use of removed records. For example, finding aids, such as inventory, register, description, isn't prepared for user. A user may usually not understand an accurate context of their creation, preservation removal. Thirdly, almost institution has arranged the removed records among other materials and books. Each institution has a different classification criteria, too. These conditions seem to make a difficulty in national management. This study investigated how the materials for Korea history which has been removed from foreign countries should be managed for effective preservation and use. For these purpose, we investigated the present conditions of removal and management of these materials. Also, we investigated examples of the foreign state, U.S.A. and Japan to have a idea for removal and management policy. Then, This study propose a establishment of information network system as management policy. To administrate removing and managing records effectively, an idea that records belong to the public should be accepted commonly. The policy for management of the record is followings. Firstly, unification of classification is necessary. The records which were removed from foreign countries had better to be classified according to their origins of states and institutions. Secondly, Finding aids should be prepared for user. There are many accession aids such as catalog, register, inventory for user in archives. This will be a efficient method for management. When a record is removed, a card for description of contents must be composed. Thirdly, a digital input and network system establishment is necessary. It will help to manage a condition of removed records. A digital input and management system establishment is not just a project of digitalizing records. At present, various finding aids need to be introduced to make a database for records and archives. An search using only 'search engine' may make a difficulty in finding materials because such method is apt to have too much or little result. So, classification, arrangement, description, response-service should be integrated in these system. The recent technical advancement give a opportunity to realize this idea. Above all, these procedure should be archival process.

• Journal of the Korean Society of Earth Science Education
• /
• v.3 no.2
• /
• pp.124-131
• /
• 2010

The purpose of this study is examine relationship of attribution styles and attitude toward Science and Science Process Skills of Science-gifted, to understand unique characteristics of the Science-gifted and to give useful information that can be use in develop special programs for the Science-gifted. The result of this study were as follows: First, there was no difference between genders. But there was a significant difference in attribution of luck. Second, there was a correlation between internal tendencies and Scientific attitude. Especially attribution of effort correlated with sub - constituent of Scientific attitude. Internal tendencies correlated with the Cognition in Scientific Professions that sub - constituent of the attitude toward Science. Third, There was a correlation between external tendencies and the interest activity in Science that sub - constituent of the attitude toward Science. There are correlations between sub - constituent of the attitude toward Science and sub - constituent of external tendencies that attribution of luck and interest in Scientific Professions, attribution of task difficulty and Cognition activity in Science. Fourth There was no correlation between Attribution styles and Science Process Skills. But Internal tendencies correlated with classification that sub - constituent of Science Process Skills. And classification correlated with attribution of ability that sub - constituent of external tendencies. Attribution of effort that sub - constituent of internal tendencies correlated with Science Process Skills.

• Atmosphere
• /
• v.21 no.3
• /
• pp.221-228
• /
• 2011

The nonlinear characteristics of summer monsoon intraseasonal oscillation (ISO) in precipitation, which is manifested as fluctuations in convection and circulation, is one of the major difficulty on the prediction of East Asian summer monsoon (EASM). The present study aims to identify the spatial distribution and time evolution of nonlinear phases of monsoon ISO. In order to classify the different phases of monsoon ISO, Self-Organizing Map(SOM) known as a nonlinear pattern recognition technique is used. SOM has a great attractiveness detecting self-similarity among data elements by grouping and clustering such self-similar components. The four important patterns are demonstrated as Meiyu-Baiu, Changma, post-Changma, and dry-spell modes. It is found that SOM well captured the formation of East Asian monsoon trough during early summer and its northward migration together with enhanced convection over subtropical western Pacific and regionally intensive precipitation including Meiyu, Changma and Baiu. The classification of fundamental large scale spatial pattern and evolutionary history of nonlinear phases of monsoon ISO provides the source of predictability for extended-range forecast of summer precipitation.