• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.839-844
• /
• 2003

This paper presents a method to segment wafer ID marks on poor quality images under uncontrolled lighting conditions of the semiconductor process. The active multiple templates matching method is suggested to search ID areas on wafers and segment them into meaningful regions and it would have been impossible to recognize characters using general OCR algorithms. This active template model is designed by applying a snake model that is used for active contour tracking. Active multiple template model searches character areas and segments them into single characters optimally, tracking each character that can vary in a flexible manner according to string configurations. Applying active multiple templates, the optimization of the snake energy is done using Greedy algorithm, to maximize its efficiency by automatically controlling each template gap. These vary according to the configuration of character string. Experimental results using wafer images from real FA environment are presented.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.3 s.35
• /
• pp.229-236
• /
• 2005

The Semantic Web enables machines to achieve an effective retrieval, integration, and reuse of web resources. The keyword search method currently used has a limit to accurate search results because of a simple string matching method in web environment. This paper proposes an Ontology-Based Information Retrieval which can solve the problems and retrieve better search results through semantic relations. In this system, we implemented the Cultural Assets Ontology based on OWL with RDQL and Jena API. we also suggest a method to handle properties stored in a database.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.518-528
• /
• 2016

The searching of an extensive document database for documents that are locally similar to a given query document, and the subsequent detection of similar regions between such documents, is considered as an essential task in the fields of information retrieval and data management. In this paper, we present a framework for such a task. The proposed framework employs the method of short-read mapping, which is used in bioinformatics to reveal similarities between genomic sequences. In this paper, documents are considered biological objects; consequently, edit operations between locally similar documents are viewed as an evolutionary process. Accordingly, we are able to apply the method of evolution tracing in the detection of similar regions between documents. In addition, we propose heuristic methods to address issues associated with the different stages of the proposed framework, for example, a frequency-based fragment ordering method and a locality-aware interval aggregation method. Extensive experiments covering various scenarios related to the search of an extensive document database for documents that are locally similar to a given query document are considered, and the results indicate that the proposed framework outperforms existing methods.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.1
• /
• pp.1-11
• /
• 1998

This paper describes design and implementation of the system that is used to efficiently retrieve music information at a digital music library. Unlike typical music information retrieval systems, this system allows the user to sing a part of the melody through the microphone which he/she wants to find rather than using title, composer or the subject catalog to search. The system then recognizes the musical notes information through the signal processing of the sounds of the entered melody, and the intervals contour is created based on this information and used as a search pattern. By running the proposed notes string search algorithm that uses the musical notes information processed with the user input, and it produces the approximate search results. Therefore, users are able to retrieve and appreciate the music whenever he/she can sing any portion of the desired music.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2000.04a
• /
• pp.239-242
• /
• 2000

Symbiotic evolutionary algorithms are stochastic search algorithms that imitate the biological coevolution process through symbiotic interactions. In the algorithms, the fitness evaluation of an individual requires first selecting symbiotic partners of the individual. The symbiotic partner effects the change of individual's fitness and search direction. In this study we are to analyze how much partnering strategies can influence the performance of the algorithms. For this goal extensive experiments are carried out to compare the performance of partnering strategies. The NKC model and the binary string covering problem are used as the test-bed problems. The experimental results indicate that there does not exist statistically significant difference in their performance.

• Proceedings of the KIEE Conference
• /
• 2000.07a
• /
• pp.156-158
• /
• 2000

This paper presents a efficient algorithm for loss reduction of distribution system by automatic sectionalizing switch operation in distribution systems of radial type. To apply genetic algorithm to reconfiguration of distribution system, in this paper we propose the string type and efficient reconfiguration procedure. We also discuss the more elaborate search techniques of solution space as well as the simple genetic algorithm. The experimental results show that the proposed genetic algorithm have the ability to search a good solution.

• Journal of Korean Library and Information Science Society
• /
• v.21
• /
• pp.253-289
• /
• 1994

The purpose of this study is about the search pattern of LINNET (Library Information Network System) OPAC users by transaction log, maintained by POSTECH(Pohang University of Science and Technology) Central Library, to provide feedback information of OPAC system design. The results of this study are as follows. First, for the period of this analysis, there were totally 11, 218 log-ins, 40, 627 transaction logs and 3.62 retrievals per a log-in. Title keyword was the most frequently used, but accession number, bibliographic control number or call number was very infrequently used. Second, 47.02% of OPAC, searches resulted in zero retrievals. Bibliographic control number was the least successful search. User displayed 2.01% full information and 64.27% local information per full information. Third, special or advanced retrieval features are very infrequently used. Only 22.67% of the searches used right truncation and 0.71% used the qualifier. Only 1 boolean operator was used in every 22 retrievals. The most frequently used operator is 'and (&)' with title keywords. But 'bibliographical control number (N) and accessionnumber (R) are not used at all with any operators. The causes of search failure are as follows. 1. The item was not used in the database. (15, 764 times : 79.42%). 2. The wrong search key was used. (3, 761 times : 18.95%) 3. The senseless string (garbage) was entered. (324 times : 1.63%) On the basis of these results, some recommendations are suggested to improve the search success rate as follows. First, a n.0, ppropriate user education and online help function let users retrieve LINNET OPAC more efficiently. Second, several corrections of retrieval software will decrease the search failure rate. Third, system offers right truncation by default to every search term. This methods will increase success rate but should considered carefully. By a n.0, pplying this method, the number of hit can be overnumbered, and system overhead can be occurred. Fourth, system offers special boolean operator by default to every keyword retrieval when user enters more than two words at a time. Fifth, system assists searchers to overcome the wrong typing of selecting key by automatic korean/english mode change.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.8
• /
• pp.43-52
• /
• 2013

The purpose of this paper is researching musical elements to analyze plagiarism between two sources. We first search digital music elements to analyze music source and examine how to use these in plagiarism analysis using compiler techniques. In addition we are used open source Java API JFugue to process complex MIDI music data simply. Therefore we designed music plagiarism analysis system by using MusicString which is supported in JFugue and construct AST after investigate MusicString's syntax processing elements to manipulate music plagiarism analysis efficiently. So far music plagiarism analysis is evaluated emotionally and subjectively. But this paper suggests first step to build plagiarism analysis systemically. If this research is well utilized, this is very meaningful to standardize systemically which music is plagiarized or not.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.10a
• /
• pp.361-367
• /
• 1998

The main two applications of the Genetic Algorithms(GA) are the optimization and the machine learning. Machine Learning has two objectives that make the complex system learn its environment and produce the proper output of a system. The machine learning using the Genetic Algorithms is called GA machine learning or genetic-based machine learning (GBML). The machine learning is different from the optimization problems in finding the rule set. In optimization problems, the population of GA should converge into the best individual because optimization problems, the population of GA should converge into the best individual because their objective is the production of the individual near the optimal solution. On the contrary, the machine learning systems need to find the set of cooperative rules. There are two methods in GBML, Michigan method and Pittsburgh method. The former is that each rule is expressed with a string, the latter is that the set of rules is coded into a string. Th classifier system of Holland is the representative model of the Michigan method. The classifier systems arrange the strength of classifiers of classifier list using the message list. In this method, the real time process and on-line learning is possible because a set of rule is adjusted on-line. A classifier system has three major components: Performance system, apportionment of credit system, rule discovery system. In this paper, we solve the food search problem with the learning and evolution of an artificial ant using the learning classifier system.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.8
• /
• pp.1986-1992
• /
• 2009

A user often searches a data by inputting a variant such as the abbreviation or substring of a word, or a misspelled word. The simple approach to the searching for variants is to build a variants dictionary. However, it entails enormous cost and time and can not handle variants by misspelling. Approximate searching, searching by approximate string matching, is a good approach to the searching. A problem in the approach is that it cannot handle variants by abbreviations. This paper propose a method for searching various variants including abbreviations and misspelled words, by using the trie indexing. First, this paper shows a variant matching method with the calculation of path weighted-metric. In addition, it provides variant searching algorithm to reduce the search time.