• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.2
• /
• pp.162-170
• /
• 2015

Commute time embedding involves computing the spectral decomposition of the graph Laplacian. It requires the computational burden proportional to $o(n^3)$, not suitable for large scale dataset. Many methods have been proposed to accelerate the computational time, which usually employ the Nystr${\ddot{o}}$m methods to approximate the spectral decomposition of the reduced graph Laplacian. They suffer from the lost of information by dint of sampling process. This paper proposes to reduce the errors by approximating the spectral decomposition of the graph Laplacian using that of the affinity matrix. However, this can not be applied as the data size increases, because it also requires spectral decomposition. Another method called approximate commute time embedding is implemented, which does not require spectral decomposition. The performance of the proposed algorithms is analyzed by computing the commute time on the patch graph.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.7
• /
• pp.29-37
• /
• 2020

In the past, meta-heuristic optimization algorithms were developed to solve the problems caused by complex nonlinearities occurring in natural phenomena, and various studies have been conducted to examine the applicability of the developed algorithms. The self-adaptive vision correction algorithm (SAVCA) showed excellent performance in mathematics problems, but it did not apply to complex engineering problems. Therefore, it is necessary to review the application process of the SAVCA. The SAVCA, which was recently developed and showed excellent performance, was applied to the advanced Muskingum flood routing model (ANLMM-L) to examine the application and application process. First, initial solutions were generated by the SAVCA, and the fitness was then calculated by ANLMM-L. The new value selected by a local and global search was put into the SAVCA. A new solution was generated, and ANLMM-L was applied again to calculate the fitness. The final calculation was conducted by comparing and improving the results of the new solution and existing solutions. The sum of squares (SSQ) was used to calculate the error between the observed and calculated runoff, and the applied results were compared with the current models. SAVCA, which showed excellent performance in the Muskingum flood routing model, is expected to show excellent performance in a range of engineering problems.

• Journal of Korea Multimedia Society
• /
• v.10 no.3
• /
• pp.411-419
• /
• 2007

Mounted on mobile device, set-top box, or digital TV, EVM is a virtual machine solution that can download and execute dynamic application programs. And the SIL(Standard Intermediate Language) is intermediate language of the EVM, which has a set of opcodes for object-oriented language and a sequential language. Since the C compiler used on each platform depends on the hardware, it converts C program to objective code, and then executes. To solve this problem, our research team developed ANSI C compiler and the EVM. Our ANSI C compiler outputs the SIL code based on stack machine. This paper presents the SIL-to-C decompiler in which converts the SIL code to three address code. Thus, the decompiler allows us to verify SIL code created by ANSI C compiler, and analyze a program from C language source level.

• Journal of KIISE
• /
• v.42 no.11
• /
• pp.1303-1313
• /
• 2015

MapReduce provides high levels of system scalability and fault tolerance for large-size data processing. A MapReduce-based k-nearest-neighbor(k-NN) join algorithm seeks to produce the k nearest-neighbors of each point of a dataset from another dataset. The algorithm has been considered important in bigdata analysis. However, the existing k-NN join query-processing algorithm suffers from a high index-construction cost that makes it unsuitable for the processing of bigdata. To solve the corresponding problems, we propose a new grid-based, k-NN join query-processing algorithm. Our algorithm retrieves only the neighboring data from a query cell and sends them to each MapReduce task, making it possible to improve the overhead data transmission and computation. Our performance analysis shows that our algorithm outperforms the existing scheme by up to seven-fold in terms of the query-processing time, while also achieving high extent of query-result accuracy.

• The KIPS Transactions:PartD
• /
• v.10D no.7
• /
• pp.1103-1114
• /
• 2003

Currently, one of the most critical issues in developing the service support system for various spatio-temporal applications is the discoverying of meaningful knowledge from the large volume of moving object data. This sort of knowledge refers to the spatiotemporal moving pattern. To discovery such knowledge, various relationships between moving objects such as temporal, spatial and spatiotemporal topological relationships needs to be considered in knowledge discovery. In this paper, we proposed an efficient method, MPMine, for discoverying spatiotemporal moving patterns. The method not only has considered both temporal constraint and spatial constrain but also performs the spatial generalization using a spatial topological operation, contain(). Different from the previous temporal pattern methods, the proposed method is able to save the search space by using the location summarization and generalization of the moving object data. Therefore, Efficient discoverying of the useful moving patterns is possible.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.2
• /
• pp.147-153
• /
• 2017

Efficient texture representation is very important in computer vision fields. The performance of texture classification or/and segmentation can be improved based on efficient texture representation. Gabor filter is a representation method that has long history for texture representation based on multi-scale analysis. Gabor filter shows good performance in texture classification and segmentation but requires much processing time. In this paper, we propose new texture representation method that is also based on multi-scale analysis. The proposed representation can provide similar performance in texture classification but can reduce processing time against Gabor filter. Experimental results show good performance of our method.

• The KIPS Transactions:PartD
• /
• v.16D no.1
• /
• pp.113-122
• /
• 2009

Since the traditional pattern mining methods only probe unspecified moving patterns that seem to satisfy users' requests among diverse patterns within the limited scopes of time and space, they are not applicable to problems involving the mining of optimal moving patterns, which contain complex time and space constraints, such as 1) searching the optimal path between two specific points, and 2) scheduling a path within the specified time. Therefore, in this paper, we illustrate some problems on mining the optimal moving patterns with complex time and space constraints from a vast set of historical data of numerous moving objects, and suggest a new moving pattern mining method that can be used to search patterns of an optimal moving path as a location-based service. The proposed method, which determines the optimal path(most frequently used path) using pattern frequency retrieved from historical data of moving objects between two specific points, can efficiently carry out pattern mining tasks using by space generalization at the minimum level on the moving object's location attribute in consideration of topological relationship between the object's location and spatial scope. Testing the efficiency of this algorithm was done by comparing the operation processing time with Dijkstra algorithm and $A^*$ algorithm which are generally used for searching the optimal path. As a result, although there were some differences according to heuristic weight on $A^*$ algorithm, it showed that the proposed method is more efficient than the other methods mentioned.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.12
• /
• pp.623-634
• /
• 2016

Most of existing frequent pattern mining methods address time efficiency and greatly rely on the primary memory. However, in the era of big data, the size of real-world databases to mined is exponentially increasing, and hence the primary memory is not sufficient enough to mine for frequent patterns from large real-world data sets. To solve this problem, there are some researches for frequent pattern mining method based on disk, but the processing time compared to the memory based methods took very time consuming. There are some researches to improve scalability of frequent pattern mining, but their processes are very time consuming compare to the memory based methods. In this paper, we present PPFP as a novel disk-based approach for mining frequent itemset from big data; and hence we reduced the main memory size bottleneck. PPFP algorithm is based on FP-growth method which is one of the most popular and efficient frequent pattern mining approaches. The mining with PPFP consists of two setps. (1) Constructing an IFP-tree: After construct FP-tree, we assign index number for each node in FP-tree with novel index numbering method, and then insert the indexed FP-tree (IFP-tree) into disk as IFP-table. (2) Mining frequent patterns with PPFP: Mine frequent patterns by expending patterns using stack based PUSH-POP method (PPFP method). Through this new approach, by using a very small amount of memory for recursive and time consuming operation in mining process, we improved the scalability and time efficiency of the frequent pattern mining. And the reported test results demonstrate them.