• The KIPS Transactions:PartB
• /
• v.13B no.2 s.105
• /
• pp.127-132
• /
• 2006

K-NN (k-Nearest Neighbors), which is a well-known instance-based learning algorithm, simply stores entire training patterns in memory, and uses a distance function to classify a test pattern. K-NN is proven to show satisfactory performance, but it is notorious formemory usage and lengthy computation. Various studies have been found in the literature in order to minimize memory usage and computation time, and NGE (Nested Generalized Exemplar) theory is one of them. In this paper, we propose RPA (Recursive Partition Averaging) and IRPA (Incremental RPA) which is an incremental version of RPA. RPA partitions the entire pattern space recursively, and generates representatives from each partition. Also, due to the fact that RPA is prone to produce excessive number of partitions as the number of features in a pattern increases, we present IRPA which reduces the number of representative patterns by processing the training set in an incremental manner. Our proposed methods have been successfully shown to exhibit comparable performance to k-NN with a lot less number of patterns and better result than EACH system which implements the NGE theory.

• Fire Science and Engineering
• /
• v.28 no.5
• /
• pp.44-51
• /
• 2014

In this study, the Mock-up office space experiments have been performed for the fire behavior analysis of the compartmented space used for the performance-based fire safety design of buildings. Mock-up test was conducted using the compartmented office space dimensions, which are 2.4 m wide, 3.6 m wide, and 2.4 m hight. Test was conducted with the combustible materials such as a desk, a chair, a computer ect. The fire load in the Mock-up office space was $18.74kg/m^2$. As a result, the temperature of the central compartment space to reach $600^{\circ}C$ were 394 to 408 s. The temperature of the corner near the entrance edge to reach $600^{\circ}C$ were 404 to 420 s. At this study, the temperature curve in the compartmented space has been predicted using the temperature data appling the BFD curve. The BFD curve factor based on the fire tests was determined by the maximum temperature of $900^{\circ}C$, 7 min to reach the maximum temperature, and the shape coefficient of 1.5. The initiating fire was rapidly increased to 9 min, and decreased.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.3
• /
• pp.297-303
• /
• 2009

In this paper, we propose the biomedical spectral pattern classification techniques by the fusion scheme based on the SpPCA and MLP in extended feature space. A conventional PCA technique for the dimension reduction has the problem that it can't find an optimal transformation matrix if the property of input data is nonlinear. To overcome this drawback, we extract features by the SpPCA technique in extended space which use the local patterns rather than whole patterns. In the classification step, individual classifier based on MLP calculates the similarity of each class for local features. Finally, biomedical spectral patterns is classified by the fusion scheme to effectively combine the individual information. As the simulation results to verify the effectiveness, the proposed method showed more improved classification results than conventional methods.

• The Journal of the Korea Contents Association
• /
• v.21 no.7
• /
• pp.157-163
• /
• 2021

A single write operation in a file system can modify multiple data, but these changes in the file system are not atomically written to disk. Thus, for the consistency of the file system, conventional journaling guarantees crash consistency instead of sacrificing the system performance. It is known that using non-volatile memory as a journal space can alleviate performance degradation due to low latency and byte-level accessibility of non-volatile memory. However, none of the journaling techniques considering non-volatile memory provide scalability. In this paper, journal space on non-volatile memory is divided into multiple regions for scalable journaling, thus dispersing concentrated operations in one region. Second, the journal area-specific operator structure is used to accelerate data write operations to storage devices. We apply the proposed technique to JFS to evaluate it on multi-core servers equipped with high-performance storage devices. The evaluation results show that the proposed technique performs better than the existing technique of the NVM-based journaling file system.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.408-408
• /
• 2023

하천공간은 하도, 사주, 식생, 하천구조물 등에 대한 특성을 지니고 있으며, 현장조사를 통해 하천공간에 대한 자료를 분석하여 기초자료를 생산한다. 기존에는 현장에서 육안조사나 지상에서 사진촬영, 스케치방법으로 하천공간특성에 대한 조사를 수행하였으나, 지상에서 조사한 자료은 하천특성에 대한 물리적·공간적 특성을 파악하기 어렵고 자료의 활용성이 낮은 한계점이 존재한다. 이와 같은 한계를 극복하기 위해 GIS 및 RS 기술을 활용한 고도화된 첨단조사 기술 및 장비가 도입되어 활용되고 있다. 본 연구에서는 하천공간특성을 GIS 기반으로 객체지향 분류 적용 연구와 분류 항목에 따른 공간분석 연구를 수행하였다. 연구를 위한 대상지역은 섬진강권역의 지석천 유역 하류부에 위치하고 있는 지석천 친수공원을 대상으로 선정하였다. 대상지역의 고해상도 항공영상을 수집 및 정합한 후 QGIS에서 제공하는 Orfeo ToolBox(OTB)의 LSMS(Large Scale Mean-Shift) 기법으로 정합한 항공영상의 객체지향 영상분할을 실시하여 벡터 레이어를 생성하였고, 하천공간특성에 따른 항목을 선정하여 각 항목의 영역에 대한 선별을 통해 훈련데이터를 생성하였다. 훈련데이터는 랜덤 포레스트를 이용하여 각 항목에 대한 자동 분류를 확인하였으며, 하천공간특성의 정량적 평가를 위해 분류된 각 항목별 공간분석을 통해 면적, 위치정보(위도, 경도, 표고)를 산정하였다. 분석 결과, 하천공간특성을 GIS 기반의 벡터 레이어와 각 항목에 대한 정량적 분석을 통해 하천공간의 DB를 구축하였다. 이와 같이 하천공간 DB 구축을 통해 전국 하천관리체계를 위한 기초자료를 구축하고자 하였다.

• Journal of Internet Computing and Services
• /
• v.15 no.2
• /
• pp.33-39
• /
• 2014

How to transfer spatial data from server to client in wireless broadcasting environment is shown as following: A server arranges data information that client wants and transfers data by one-dimensional array for broadcasting cycle. Client listens data transferred by the server and returns resulted value only to server. Recently number of users using location-based services is increasing alongside number of objects, and data volume is changing into large amount. Large volume of data in wireless broadcasting environment may increase query time of client. Therefore, we propose Client based Data Scheduling (CDS) for efficient data scheduling in wireless broadcasting environment. CDS divides map and then calculates total sum of objects for each grid by considering number of objects and data size within divided grids. It carries out data scheduling by applying hot-cold method considering total data size of objects for each grid and number of client. It's proved that CDS reduces average query processing time for client compared to existing method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.2
• /
• pp.239-250
• /
• 2022

LoRa is a physical layer technology designed to secure highly reliable long-range communication with introducing loco parentis tree network and chirp spreading spectrum. Since since a leaf can send message to more than one parents simultaneously with a single transmission in a region, packet delivery ratio increases logarithmically as the number of gateways increases. The delivery ratio, however, dramatically collapses even under loco parentis tree topology due to the limitations of ALOHA-like primitive MAC, . The proposed method is intended to exploit SDMA approach to reuse frequency in an area. With the view, TxPower of each sender for each message in a concurrent transmission is elaborately controlled to survive the collision at different gateway. Thus, the gain from the capture effect improves the capacity of resource-hungry Low Power and Wide Area Networks.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.194-198
• /
• 2021

LoRa is a physical layer technology that is designed to provide a reliable long-range communication with introducing CSS and with introducing a loco parentis tree network. Since a leaf can utilize multiple parents at the same time with a single transmission, PDR increases logarithmically as the number of gateways increases. Because of the ALOHA-like MAC of LoRa, however, the PDR degrades even under the loco parentis tree topology similarly to the single-gateway environment. Our proposed method is aimed to achieve SDMA approach to reuse the same frequency in different areas. For that purpose, it elaborately controls each TxPower of the senders for each message in concurrent transmission to survive the collision at each different gateway. The gain from this so-called capture effect increases the capacity of resource-hungry LPWAN. Compared to a typical collision-free controlled-access scheme, our method outperforms by 10-35% from the perspective of the total count of the consumed time slots. Also, due to the power control mechanism in our method, the energy consumption reduced by 20-40%.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.10b
• /
• pp.324-329
• /
• 2006

Non-negative matrix factorization(NMF) 기법은 음이 아닌 값으로 구성된 데이터를 두 종류의 양의 행렬의 곱의 형식으로 분할하는 데이터 분석기법으로서, 텍스트마이닝, 바이오인포매틱스, 멀티미디어 데이터 분석 등에 활용되었다. 본 연구에서는 기본 NMF 기법에 기반하여 텍스트 문서로부터 토픽을 추출하고 동시에 이를 가시적으로 도시하기 위한 Topographic NMF (TNMF) 기법을 제안한다. TNMF에 의한 토픽 가시화는 데이터를 전체적인 관점에서 보다 직관적으로 파악하는데 도움이 될 수 있다. TNMF는 생성모델 관점에서 볼 때, 2개의 은닉층을 갖는 계층적 모델로 표현할 수 있으며, 상위 은닉층에서 하위 은닉층으로의 연결은 토픽공간상에서 토픽간의 전이확률 또는 이웃함수를 정의한다. TNMF에서의 학습은 전이확률값의 연속적 스케줄링 과정 속에서 반복적 파리미터 갱신 과정을 통해 학습이 이루어지는데, 파라미터 갱신은 기본 NMF 기반 학습 과정으로부터 유사한 형태로 유도될 수 있음을 보인다. 추가적으로 Probabilistic LSA에 기초한 토픽 가시화 기법 및 희소(sparse)한 해(解) 도출을 목적으로 한 non-smooth NMF 기법과의 연관성을 분석, 제시한다. NIPS 학회 논문 데이터에 대한 실험을 통해 제안된 방법론이 문서 내에 내재된 토픽들을 효과적으로 가시화 할 수 있음을 제시한다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.20-26
• /
• 1994

In this paper, we propose an algorithm on partitioning a network into several subsets where the cost of a net which connects nodes in k subsets is given as k(k-1)/2 indicating the typical pattern of complete graphs. This problem is one of generalizations for multiple-way partitioning proposed by Sanchis. $^{[5]}$ Its solution can be applied to resource allocation problem in distributed systems. The proposed algorithm expanded the algorithm of Fiduccia and Mattheyses$^{[3]}$ to handle the multiple-way partitioning simultaneously. It has time and space complexity linear to the size of the network. To evaluate the performance of the proposed algorithm, we implemented also a traditional cluster growth method which groups connected nodes for nets, and compared experimental results with those of the proposed algorithm. The proposed algorithm shows some enhancement made.