• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.478-487
• /
• 2004

We had proposed the RPA(Recursive Partition Averaging) method in order to improve the storage requirement and classification rate of the Memory Based Reasoning. That algorithm worked not bad in many area, however, the major drawbacks of RPA are it's partitioning condition and the way of extracting major patterns. We propose an adaptive RPA algorithm which uses the FPD(feature-based population densimeter) to stop the ARPA partitioning process and produce, instead of RPA's averaged major pattern, optimizing resulting hyperrectangles. The proposed algorithm required only approximately 40% of memory space that is needed in k-NN classifier, and showed a superior classification performance to the RPA. Also, by reducing the number of stored patterns, it showed an excellent results in terms of classification when we compare it to the k-NN.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.3
• /
• pp.139-147
• /
• 2009

The memory based reasoning just stores in the memory in the form of the training pattern of the representative pattern. And it classifies through the distance calculation with the test pattern. Because it uses the techniques which stores the training pattern whole in the memory or in which it replaces training patterns with the representative pattern. Due to this, the memory in which it is a lot for the other machine learning techniques is required. And as the moreover stored training pattern increases, the time required for a classification is very much required. In this paper, We propose the EAS(Evaluation And Selection) algorithm in order to minimize memory usage and to improve classification performance. After partitioning the training space, this evaluates each partitioned space as MDL and PM method. The partitioned space in which the evaluation result is most excellent makes into the representative pattern. Remainder partitioned spaces again partitions and repeat the evaluation. We verify the performance of Proposed algorithm using benchmark data sets from UCI Machine Learning Repository.

• Journal of Korea Spatial Information System Society
• /
• v.8 no.2 s.17
• /
• pp.39-52
• /
• 2006

With the recent the use of spatio-temporal data mining which can extract various knowledge such as movement patterns of moving objects in history data of moving object gets increasing. However, the existing movement pattern extraction methods create lots of candidate movement patterns when the minimum support is low. Therefore, in this paper, we suggest the STMPE(Spatio-Temporal Movement Pattern Extraction) algorithm in order to efficiently extract movement patterns of moving objects from the large capacity of spatio-temporal data. The STMPE algorithm generalizes spatio-temporal and minimizes the use of memory. Because it produces and keeps short-term movement patterns, the frequency of database scan can be minimized. The STMPE algorithm shows more excellent performance than other movement pattern extraction algorithms with time information when the minimum support decreases, the number of moving objects increases, and the number of time division increases.

• Review of KIISC
• /
• v.18 no.5
• /
• pp.62-72
• /
• 2008

인터넷 뱅킹은 인터넷을 통해 금융 업무를 처리하는 시스템으로, 시 공간적 제약이 없어 이용자가 크게 증가하고 있지만 인터넷을 기반으로 한 웹 공격으로 인하여 많은 위협을 받고 있다. 인터넷 뱅킹은 서비스를 제공하는 은행에 따라 사용자 인터페이스와 처리 방법이 매우 다양하므로, 인터넷 뱅킹 시스템을 목표로 한 웹 공격을 탐지하기 위해서는 해당 인터넷 뱅킹 서비스의 특징을 반영할 수 있는 고유의 패턴을 생성해야 한다. 본 논문에서는 서열 정렬 알고리즘을 이용하여 인터넷 뱅킹 이용에 대한 정상 및 비정상 패턴을 자동으로 생성하여 웹 공격을 탐지하고 분석하는 방법을 제안한다. 제시한 방법의 성능 평가를 위하여, 모의 인터넷 뱅킹 프로그램을 설치한 후 정상적인 이용과 웹 공격을 시도한 자료를 구분하여 수집하고 유사도를 측정하였다. 실험결과 제안된 기법이 오탐율이 낮고 탐지 성능 또한 뛰어남을 확인하였다. 그리고 전문가의 도움 없이 정상 패턴과 비정상 패턴을 생성할 수 있어 효율적으로 변형된 공격이나 새로운 공격을 차단하고 비정상 행위에 판단에 대한 근거를 제시할 수 있음을 보였다.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.7
• /
• pp.1849-1857
• /
• 1999

We proposed the RPA (Recursive Partition Averaging) method in order to improve the storage requirement and classification rate of the Memory Based Reasoning. This algorithm recursively partitions the pattern space until each hyperrectangle contains only those patterns of the same class, then it computes the average values of patterns in each hyperrectangle to extract a representative. Also we have used the mutual information between the features and classes as weights for features to improve the classification performance. The proposed algorithm used 30~90% of memory space that is needed in the k-NN (k-Nearest Neighbors) classifier, and showed a comparable classification performance to the k-NN. Also, by reducing the number of stored patterns, it showed an excellent result in terms of classification time when we compare it to the k-NN.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.25 no.2
• /
• pp.59-71
• /
• 2022

This study considered regional differences in the university-industry collaboration of Korean universities and performed cluster analysis to identify the spatial range with high university-industry collaboration connectivity. By university establishment type, it was found that the university-industry collaboration capacity of the major national university was superior overall, especially in the technology transfer & commercialization sector and the infrastructure sector, compared to private universities and general national universities. The spatial pattern of university-industry collaboration capacity showed relatively clear differences by city and province. In terms of university-industry collaboration capacity by sector, it was confirmed that the regional gap was not large in the talent training sector and the infrastructure sector, but the regional gap was relatively large in the technology transfer & commercialization sector and the start-up sector. As a result of the cluster analysis to identify a spatial range with high connectivity in terms of similarity and spatial proximity of university-industry collaboration patterns, it is divided into 15 clusters. It is found that most of major national universities are included in one of 15 clusters where all sectors of university-industry collaboration are strong. Therefore, as a policy measure to achieve regional innovative growth through enhancing the effectiveness of university-industry collaboration, we propose the establishment of a hub & spoke network-type collaboration system in which a major national university acts as a hub and nearby local universities play a spoke role.

• Journal of the Korean association of regional geographers
• /
• v.21 no.1
• /
• pp.114-136
• /
• 2015

Manufacturing FDI is mainly conducted by establishing production firms in the host country. The foreign branch facilities which are closely connected to headquarters in the home countries through the global supply chain have decisive effects on the integration of global production system and on the pattern changes in international logistics. Therefore, the study on the development process of manufacturing supply chain through FDI can be the key factor to understand the global production and international logistics system. This study analyzes the spatial and temporal patterns of direct investments in China by Korean companies in the manufacturing sectors. The results show a great deal of differential in periods and regions in investment. At the early period of Korean manufacturing FDI, the coast provinces in China which are adjacent to Korean peninsula were main investment target regions. However recently, coastal and inland provinces far away from Korea have been important for attracting Korean manufacturing capital. If further affiliated studies are done, the results of this study would be more beneficial to understand the logistics flows of regional supply chain based on the manufacturing FDI.

• The Journal of the Korea Contents Association
• /
• v.12 no.4
• /
• pp.95-107
• /
• 2012

Communication skill to maintain personal relationships is a crucial factor in our social life. However, even though a number of people are experiencing difficulty in communication owing to their poor interpersonal interchanges in real life, they are interacting with virtual characters in artificially created virtual space. In order to remedy such a lack of communication among game players, first, they need to search appropriate ways to communicate each other in real life, but not in virtual space, or to create a game space as a new place for interaction among them by utilizing advantages of the game world. To accomplish such an aim, it is necessary to precede the research on what kind of relations or characteristics for game players' interaction to others in the real life and game world. For the reason, in this paper, we investigate interaction patterns of game players in real life and in the virtual space. In order to perform this investigation, an ego-gram, which demonstrates the pattern of the ego states as a field-oriented approach is used as the research method. The result shows that there are differences between patterns in each spaces. so, we apply a factor analysis and analyse the relationship between the transaction pattern of players in two spaces using the ego-gram.

• Broadcasting and Media Magazine
• /
• v.18 no.3
• /
• pp.27-39
• /
• 2013

디스플레이 칩 상에 홀로그램을 표시할 때는 칩 자체의 화소 배열 구조에 의해 주어지는 회절 현상에 의해 표시되는 홀로그램의 재생상이 위치될 상 공간과 이 재생상을 볼 수 있는 시역이 정의된다. 그리고 상기 회절 현상뿐만 아니라 홀로그램 자체의 간섭무늬 패턴 구조에 의한 회절에 의해 생성된 회절 패턴이 상 공간 내에서 재생상과 중첩되어 나타나므로 이들은 잡음으로 주어진다. 이들 회절 패턴들은 칩 자체의 화소 구조 및 간섭무늬의 특성에 의해 주어지므로, 이들 칩이나 평판 표시패널의 고유한 것이어서 제거가 아주 어렵다. 그러므로 이들 칩들을 홀로그램 표시장치로 지속적인 사용을 위해서는 이들 문제의 해결이 필요하다.

• Journal of the Korean Regional Science Association
• /
• v.34 no.2
• /
• pp.35-47
• /
• 2018

This research aims to examine the spatial patterns of vacant housings and the factors determining housing vacancy in Daegu using individual-level building DB. The results of the spatial pattern analysis showed a donut shaped-spatial concentration of vacant housings in the central areas of the city. The results of logistic regression analysis revealed that not only individual building characteristics, such as building area, number of floors, and building age, but also socio-economic characteristics of community, such as urban redevelopment district, number of adjacent vacancies, recent population change, and ratio of elderly, are important factors affecting housing vacancies.