• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.3
• /
• pp.217-222
• /
• 2015

Clustering is one of important data mining techniques known as exploratory data analysis and is being applied in various engineering and scientific fields such as pattern recognition, remote sensing, and so on. The method organizes data by abstracting underlying structure either as a grouping of individuals or as a hierarchy of groups. Weather radar observes atmospheric objects by utilizing reflected signals and stores observed data in corresponding coordinate. To analyze the radar data, it is needed to be separately organized precipitation and non-precipitation echo based on similarities. Thus, this paper studies to apply clustering method to radar data. In addition, in order to solve the problem when precipitation echo locates close to non-precipitation echo, fuzzy logic based clustering method which can consider both distance and other properties such as reflectivity and Doppler velocity is suggested in this paper. By using actual cases, the suggested clustering method derives better results than previous method in near-located precipitation and non-precipitation echo case.

• Journal of Internet Computing and Services
• /
• v.13 no.1
• /
• pp.65-74
• /
• 2012

Cloud services are used for improving business. Moreover, customer relationship management(CRM) approaches use social networking as tools to enhance services to customers. However, most cloud systems do not support the semantic structures, and because of this, vital information from social network sites is still hard to process and use for business strategy. This paper proposes a collaboration framework based on social semantic web for cloud system. The proposed framework consists of components to support social semantic web to provide an efficient collaboration system for cloud consumers and service providers. The knowledge acquisition module extracts rules from data gathered by social agents and these rules are used for collaboration and business strategy. This paper showed the implementations of processing of social network site data in the proposed semantic model and pattern extraction which was used for the virtual grouping of cloud service providers for efficient collaboration.

• Information Systems Review
• /
• v.10 no.2
• /
• pp.195-209
• /
• 2008

Customers' needs change every moment. Profitability of stores can't be increased anymore with an existing standardized chain store management. Accordingly, a personalized store management tool needs through prediction of customers' preference. In this study, we propose a recommending procedure using dynamic customers' preference by analyzing the transaction database. We utilize self-organizing map algorithm and association rule mining which are applied to cluster the chain stores and explore purchase sequence of customers. We demonstrate that the proposed methodology makes an effect on recommendation of products in the market which is characterized by a fast fashion and a short product life cycle.

• Economic and Environmental Geology
• /
• v.31 no.4
• /
• pp.277-290
• /
• 1998

The purpose of this research is to investigate the dispersion pattern of gold during skarnization and genesis of gold mineralization in the Sangdong skarn deposits. The Sangdong scheelite orebodies are embedded in the Cambrian Pungchon Limestone and limestone interbedded in the Myobong Slate of the Cambrian age. The tungsten deposits are classified as the Hangingwall Orebody, the Main Orebody and the Footwall Orebody as their stratigraphic locations. Recently, the Sangdong granite of the Cretaceous age (85 Ma) were found by underground exploratory drillings below the orebodies. In geochemisty, the W, Mo, Bi and F concentrations in the granite are significantly higher than those in the Cretaceous granitoids in southern Korea. Highest gold contents are associated with quartz-hornblende skarn in the Main Orebody and pyroxene-hornblende skarn in the Hangingwall Orebody. Also Au contents are closely related to Bi contents. This could be inferred that Au skarns formed from solutions under reduced environment at a temperature of $270^{\circ}C$. According to the multiple regression analysis, the variation of Au contents in the Main Orebody can be explained (87.5%) by Ag, As, Bi, Sb, Pb, Cu. Judging from the mineralogical, chemical and isotope studies, the genetic model of the deposits can be suggested as follows. The primitive Sangdong magma was enriched in W, Mo, Au, Bi and volatiles (metal-carriers such as $H_2O$, $CO_2$ and F). During the upward movement of hydrothermal ore solution, the temperature was decreased, and W deposits were formed at limestone (in the Myobong Slate and Pungchon Limestone). In addition, meteoric water influx gave rise to the retrogressive alterations and maximum solubility of gold, and consequently higher grade of Au mineralization was deposited.

• Journal of Internet Computing and Services
• /
• v.5 no.5
• /
• pp.93-103
• /
• 2004

The main function of the intrusion Detection System (IDS) usee to be more or less passive detection of the intrusion evidences, but recently it is developed with more diverse types and methodologies. Especially, it is required that the IDS should process large system audit data fast enough. Therefore the data mining or neural net algorithm is being focused on, since they could satisfy those situations. In this study, we first surveyed and analyzed the several recent intrusion trends and types. And then we designed and implemented an IDS using back-propagation algorithm of the neural net, which could provide more effective solution. The distinctive feature of our study could be stated as follows. First, we designed the system that allows both the Anomaly dection and the Misuse detection. Second, we carried out the intrusion analysis experiment by using the reliable KDD Cup ‘99 data, which would provide us similar results compared to the real data. Finally, we designed the system based on the object-oriented concept, which could adapt to the other algorithms easily.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.106-114
• /
• 2009

In this paper we proposed a fast method for a K-Means Clustering algorithm. The main characteristic of this method is that it uses precalculated data which possibility of change is high in order to speed up the algorithm. When calculating distance to cluster centre at each stage to assign nearest prototype in the clustering algorithm, it could reduce overall computation time by selecting only those data with possibility of change in cluster is high. Calculation time is reduced by using the distance information produced by K-Means algorithm when computing expected input data whose cluster may change, and by using such distance information the algorithm could be less affected by the number of dimensions. The proposed method was compared with original K-Means method - Lloyd's and the improved method KMHybrid. We show that our proposed method significantly outperforms in computation speed than Lloyd's and KMHybrid when using large size data which has large amount of data, great many dimensions and large number of clusters.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.129-130
• /
• 2018

Big data analysis is a technique for effectively analyzing unstructured data such as the Internet, social network services, web documents generated in the mobile environment, e-mail, and social data, as well as well formed structured data in a database. The most big data analysis techniques are data mining, machine learning, natural language processing, and pattern recognition, which were used in existing statistics and computer science. Global research institutes have identified analysis of big data as the most noteworthy new technology since 2011. Therefore, companies in most industries are making efforts to create new value through the application of big data. In this study, we analyzed using the Social Matrics which a big data analysis tool of Daum communications. We analyzed public perceptions of "Artificial Intelligence" keyword, one month as of May 19, 2018. The results of the big data analysis are as follows. First, the 1st related search keyword of the keyword of the "Artificial Intelligence" has been found to be technology (4,122). This study suggests theoretical implications based on the results.

• Journal of the Korea Convergence Society
• /
• v.9 no.7
• /
• pp.1-7
• /
• 2018

Feature selection, one of data preprocessing techniques, is one of major research areas in many applications dealing with large dataset. It has been used in pattern recognition, machine learning and data mining, and is now widely applied in a variety of fields such as text classification, image retrieval, intrusion detection and genome analysis. The proposed method is based on a genetic algorithm which is one of meta-heuristic algorithms. There are two methods of finding feature subsets: a filter method and a wrapper method. In this study, we use a wrapper method, which evaluates feature subsets using a real classifier, to find an optimal feature subset. The training dataset used in the experiment has a severe class imbalance and it is difficult to improve classification performance for rare classes. After preprocessing the training dataset with SMOTE, we select features and evaluate them with various machine learning algorithms.

• Geomechanics and Engineering
• /
• v.15 no.4
• /
• pp.927-935
• /
• 2018

The mechanical behavior of the brittle material samples containing the internal and edge cracks are studied under direct shear tests. It is tried to investigate the effects of stress interactions and stress intensity factors at the tips of the pre-existing cracks on the failure mechanism of the bridge areas within these cracks. The direct shear tests are carried out on more than 30 various modeled samples each containing the internal cracks (S models) and edge cracks (E models). The visual inspection and a low power microscope are used to monitor the failure mechanisms of the tested samples. The cracks initiation, propagation and coalescences are being visualized in each test and the detected failure surfaces are used to study and measure the characteristics of each surface. These investigations show that as the ratio of the crack area to the total shear surface increases the shear failure mode changes to that of the tensile. When the bridge areas are fixed, the bridge areas in between the edge cracks have less strength than those of internal cracks. However, the results of this study show that for the case of internal cracks as the bridge area is increased, the strength of the material within the bridge area is decreased. It has been shown that the failure mechanism and fracture pattern of the samples depend on the bridge areas because as the bridge area decreases the interactions between the crack tip stress fields increases.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.5
• /
• pp.81-91
• /
• 2012

Recently, as location-determination of indoor users is available with the development of variety of localization techniques for indoor location-based service, diverse indoor location based services are proposed. Accordingly, it is necessary to develop individualized POI recommendation service for recommending most interested points of large-scale commercial spaces such as shopping malls and departments. For POI recommendation, it is necessary to study the method for exploring location which users are interested in location with considering user's mobility in large-scale commercial spaces. In this paper, we proposed POI recommendation system with the definition of users' as 'Stay point' in order to consider users' various interest locations. By using the proposed algorithm, we analysis users' Stay points, then mining the users' visiting pattern to finished the proposed. POI Recommendation System. The proposed system decreased data more dramatically than that of using user's entire mobility data and usage of memory.