• Korean Journal of Cognitive Science
• /
• v.10 no.4
• /
• pp.71-83
• /
• 1999

Conventional symbolic inference systems lack flexibility because they do not well reflect flexible semantic structure of knowledge and use symbolic logic for their basic inference mechanism. For solving this problem. we have recently proposed the 'Connectionist Semantic Network(CSN)' as a model for flexible knowledge representation and inference based on neural networks. The CSN is capable of carrying out both approximate reasoning and commonsense reasoning based on similarity and association. However. we have difficulties in representing general and structured high-level knowledge and variable binding using the connectionist framework of the CSN. In this paper. we propose a hybrid system called SymCSN(Symbolic CSN) that combines a symbolic module for representing general and structured high-level knowledge and a connectionist module for representing and learning low-level semantic structure Simulation results show that the SymCSN is a plausible model for human-like flexible knowledge representation and inference.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.6B
• /
• pp.1032-1041
• /
• 1999

This paper proposes on-line scheduling algorithms that reduce the largest weighted error incurred by preemptive imprecise tasks running on a single processor system. The first one is a two-level algorithm. The top-level scheduling, which is executed whenever a new task arrives, determines the processing times to be allotted to tasks in such a way to minimize maximum weighted error as well as to minimize total error. The lower-level algorithm actually allocates the processor to the tasks. The second algorithm extends the on-line algorithm studied by Shih and Liu[4] by formalizing the top-level algorithm mathematically. The numerical simulation shows that the proposed algorithm outperforms the previous works in the sense that it greatly reduces the largest weighted error.

• Journal of Korean Library and Information Science Society
• /
• v.36 no.4
• /
• pp.309-328
• /
• 2005

Comparative analyses were carried out on SCI publication results produced by scientists during their employment at the five domestic ocean-related college and six major oceanographic institutes in other countries. Assessment of research performance included bibliometric indicators : type of research categories, and, evaluation of the reliability of citation data to establish future research goals. Analysis parameters also included number of SCI publication results, cited frequency, higher ranked author group, higher ranked Journal group, higher ranked SCI category for each academic and research Institutes in oceanography. Such analyses can help improve and achieve institutional goals in the future as well as help managing and prioritizing majors scientific journals and information services.

• Journal of Korean Library and Information Science Society
• /
• v.42 no.4
• /
• pp.281-309
• /
• 2011

This research designed and proposed metadata elements for the long-term preservation of digital research outcomes collected and managed in electronic file form. Metadata structure of both domestic and foreign agencies that manage research outcomes and CERIF and PREMIS-preserved preservation metadata were referred to design the metadata elements. The metadata design was divided into research information, bibliographic information and archival information. Research information is divided into project, name and institution, and composed of eight top-level elements and 29 sub-level elements. Bibliographic information is composed of 13 top-level elements as the source of research reports. Archival information is composed of 18 top-level elements and 62 sub-level elements as the elements for context, structure, contents and management process of the electronic records.

• Journal of Educational Research in Mathematics
• /
• v.25 no.3
• /
• pp.303-321
• /
• 2015

This research purposes to find out the mathematical cognitive characteristics of Korean students and compare it with that of TIMSS 2011 high-achieving countries based on the Cognitive Diagnostic Theory. Based on framework and questions of TIMSS 2011, we select cognitive attributes. Using the data of 8th grade students' mathematical achievement in TIMSS 2011, we compare and analyze the top 15-countries students' cognitive traits. As a result, cognition domain of TIMSS 2011 is reclassified as 9 cognitive attributes. we could distinguish between easy attributes and difficult attributes that students in each country relatively think. Especially, Students of Korea relatively think Recall/Recognize, Compute, Classify/Measure and Represent are easy. On the other hand, relatively they have difficulties in Retrieve, Implement, and Generalize. Based on this research result, It is necessary to establish an educational measures for each attributes which students have difficulties.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.1
• /
• pp.76-81
• /
• 2009

A novel design method of time-interleaved subranging ADC is presented. We use the bisection method to let only half of comparators in typical subranging ADC working in every clock cycle. Thus, we are able to reduce the number of comparators by half. It is possible to reduce the die size. An example of 8-bit time-interleaved subranging ADC operates at 40MHz sampling rate and 1.8V supply voltage is demonstrated. The power consumption of the proposed circuit is only 10mV with SPECTRE simulation. Compared with the typical subranging ADC, our bisection method is able to reduce up to 40% in die size.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.9
• /
• pp.139-147
• /
• 2014

In this paper, we present a novel method for simultaneous detection and tracking of multiple objects using dual-layer particle filtering. The proposed dual-layer particle sampling (DLPS) algorithm consists of parent-particles (PP) in the first layer for detecting multiple objects and child-particles (CP) in the second layer for tracking objects. In the first layer, PPs detect persons using a classifier trained by the intersection kernel support vector machine (IKSVM) at each particle under a randomly selected scale. If a certain PP detects a person, it generates CPs, and makes an object model in the detected object region for tracking the detected object. While PPs that have detected objects generate CPs for tracking, the rest of PPs still move for detecting objects. Experimental results show that the proposed method can automatically detect and track multiple objects, and efficiently reduce the processing time using the sampled particles based on motion distribution in video sequences.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.45-52
• /
• 2017

This paper presents a high-performance software-hardware dehazing system based on a dedicated hardware accelerator for the haze removal. In the proposed system, the dedicated hardware accelerator performs the dark-channel-prior-based dehazing process, and the software performs the other control processes. For this purpose, the dehazing process is realized as an OpenCL kernel by finding the inherent parallelism in the algorithm and is synthesized into a hardware by employing a high-level-synthesis technique. The proposed system executes the dehazing process much faster than the previous software-only dehazing system: the performance improvement is up to 96.3% in terms of the execution time.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.4
• /
• pp.715-725
• /
• 2014

This study uses a quantile regression analysis to investigate intergenerational economic mobility in Korea. The analysis is based on data from the 1st through 11th waves of the Korean Labor and Income Panel Study (KLIPS) conducted from 1998-2008. The household nature of the data allows us to link parents' incomes to children's incomes at different points in time. Using a quantile regression analysis instead of mean one reveals that the effect of fathers' earnings are different across the conditional distribution of sons' earnings, particularly being larger on the upper quantile than on the lower quantile. After controlling effect of sons' college education by including a dummy variable for the degree, however, the pattern among quantile effects for fathers' earnings is no longer clear. Instead a new pattern emerges that education has a much larger effect on the upper quantiles than on the lower ones. Using nonparametric estimates of conditional density curves based on the quantile regression results, we derive some interesting features in graphical forms, which are not obvious in numerical analysis.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.6
• /
• pp.661-667
• /
• 2005

In this paper, we propose a new hybrid genetic algorithm for sequential ordering problem (SOP). In the proposed genetic algorithm, the Voronoi quantized crossover (VQX) is used as a crossover operator and the path-preserving 3-Opt is used as a local search heuristic. VQX is a crossotver operator that uses the epistasis information of given problem instance. Since it is a crossover proposed originally for the traveling salesman problem (TSP), its application to SOP requires considerable modification. In this study, we appropriately modify VQX for SOP, and develop three algorithms, required in the modified VQX, named Feasible solution Generation Algorithm, Precedence Cycle Decomposition Algorithm, and Genic Distance Assignment Method. The results of the tests on SOP instances obtained from TSPLIB and ZIB-MP-Testdata show that the proposed genetic algorithm outperforms other genetic algorithms in stability and solution quality.