• Journal of the Korea Society of Computer and Information
• /
• v.17 no.10
• /
• pp.25-33
• /
• 2012

In this paper, we propose a rule-based model to correct errors in a speech recognition result in the mobile device environment. The proposed model considers the mobile device environment with limited resources such as processing time and memory, as follows. In order to minimize the error correction processing time, the proposed model removes some processing steps such as morphological analysis and the composition and decomposition of syllable. Also, the proposed model utilizes the longest match rule selection method to generate one error correction candidate per point, assumed that an error occurs. For the purpose of deploying memory resource, the proposed model uses neither the Eojeol dictionary nor the morphological analyzer, and stores a combined rule list without any classification. Considering the modification and maintenance of the proposed model, the error correction rules are automatically extracted from a training corpus. Experimental results show that the proposed model improves 5.27% on the precision and 5.60% on the recall based on Eojoel unit for the speech recognition result.

• Journal of KIISE:Information Networking
• /
• v.30 no.2
• /
• pp.282-292
• /
• 2003

This paper proposes an efficient data structure for forwarding IPv4 and IPv6 packets at the gigabit speed in backbone routers. The LPM(Longest Prefix Matching) search becomes a bottleneck of routers' performance since the LPM complexity grows in proportion to the forwarding table size and the address length. To speed up the forwarding process, this paper introduces a data structure named BMT(Bit-Map Tie) to minimize the frequent main memory accesses. All the necessary search computations in BMT are done over a small index table stored at cache. To build the small index table from the tie representation of the forwarding table, BMT represents a link pointer to the child node and a node pointer to the corresponding entry in the forwarding table with one bit respectively. To improve the poor performance of the conventional tries when their height becomes higher due to the increase of the address length, BMT adopts a binary search algorithm for determining the appropriate level of tries to start. The simulation experiments show that BMT compacts the IPv4 backbone routers' forwarding table into a small one less than 512-kbyte and achieves the average speed of 250ns/packet on Pentium II processors, which is almost the same performance as the fastest conventional lookup algorithms.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.167-173
• /
• 2023

The cash flow matching problem(CFMP) aims to minimize the initial investment by paying the total amount due for the T-year in principal and interest of bonds or bank deposits without paying the full amount in cash. Linear programming(LP) is the only known way to solve CFMP. The linear programming method is a problem that optimizes T linear functions, and it cannot be solved by handwriting, so LINGO, which is a solution to the linear programming method, is used. This paper proposes an algorithm that obtains the solution of CFMP solely by handwriting without the help of LINGO. The proposed algorithm determines the amount of bond purchases by covering payments until the previous year of the next maturity bond in the order that the maturity date falls from the longest to the short term. In addition, until the year before the maturity of the shortest maturity bond, the amount of deposit covered by the principal and interest of the bank deposit was determined. As a result of applying the proposed algorithm to two experimental data, it was shown that more accurate results can be obtained compared to the linear programming method.

• The Journal of Korean Association of Computer Education
• /
• v.3 no.2
• /
• pp.115-127
• /
• 2000

Korean error detectors of word processors have defects such as inconvenience that users choose one of error groups, lower detecting rate of 60%, and slow processing time. In this study, I proposed a resolution method of these defects. For these, I applied bidirectional longest match strategy for morphological analysis to improve processing time. I suggested dictionaries and several algorithms such as seperation of compound noun and assistant declinable words, correction of typing error to improve processing time and to guarantee correction accuracy. I also suggested a distinguishable method for dependent noun/suffix and Josa/Eomi where many ambiguities are generated, and a distinguishable method for Korean "로써/로서" to improve the reliability of the correction system.

• Annual Conference on Human and Language Technology
• /
• 2000.10d
• /
• pp.331-336
• /
• 2000

MATES/CK는 기계번역 시스템에서 전통적으로 사용하고 있는 세 단계(분석/변환/생성)에 의해서 중한 번역을 수행하는 시스템이다. MATES/CK는 시스템 성능을 높이기 위해 패턴 기반과 통계적 정보를 이용한다. 태거(Tagger)는 중국어 단어 분리를 최장일치법으로 수행하기 때문에 일부 단어에 대해 오류를 범하게 되고 품사(POS : Part Of Speech) 태깅 시 확률적 정보만 이용하여 특정 단어가 다 품사인 경우 그 단어에 대해 특정 품사만 태깅되는 문제점이 발생한다. 또한 중국어 및 외국어 인명 및 지명에 대한 미등록들에 대해서도 올바른 결과를 도출하지 못한다. 사전에 있어서 텍스트 기반으로 존재하여 이를 관리하기에 힘이 든다. 본 논문에서는 단어 분리 오류 및 품사 태깅 오류를 해결하기 위해 중국어 태깅 제약 규칙을 적용하는 방법을 제시하고 중국어 및 외국어 인명/지명에 대한 미등록어 처리방법을 제시한다. 또한 중국어 사전 관리에 대해 알아본다.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.10 no.2
• /
• pp.58-64
• /
• 2008

This study was carried out to determine the availability of development rate(DVR) model for predicting bud break time of 'Niitaka' pear. In addition, the study also assessed the effect of an increase in temperature on predicting bud dormancy breaking time in winter season. The DVR model, which was developed for 'Kosui' pear, illustrated that the bud break time(e.g., Dec. 12-Dec. 19) of 'Niitaka' pear observed in the field corresponded to the predicted bud break time by DVR model. This result indicates that the bud break time of 'Niitaka' pear can be predicted by DVR model tuned for 'Kosui' pear. As the temperature increased during the winter season, the bud break time was delayed. Chilling requirement deficiency for the bud break time is expected in Jeju Province when the temperature increased above $4^{\circ}C$ in winter season.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.146-147
• /
• 2012

InP 기판위에 자발성장법으로 성장된 InAs 양자점은 $1.55{\mu}m$ 영역에서 발진하는 양자점 반도체 레이저 다이오드 및 광 증폭기를 제작할 수 있기 때문에 많은 관심을 받고 있다. 광통신 대역의 $1.55{\mu}m$ 반도체 레이저 다이오드 및 광 증폭기 분야에서 InAs/InP 양자점이 많은 관심을 받고 있으나, InAs/GaAs 양자점에 비해 제작이 어려운 단점을 가지고 있다. InAs/InP 양자점은 InAs/GaAs 양자점에 비해 격자 불일치가 작아 양자점의 크기가 크고 특히 As 계 박막과 P 계박막의 계면에서 V 족 원소 교환 반응으로 계면 특성 저하가 발생하여 성장이 까다롭다. As 과 P 간의 교환반응은 성장온도와 V/III 에 의해 크게 영향을 받는 것으로 보고되었다. 그러나, P계 InGaAsP 박막 위에 InAs 성장 시 발생하는 As/P 교환반응에 대한 연구는 매우 적다. 본 연구에서는 InGaAsP 박막 위에 InAs 양자점 성장 시 GI (growth interruption)에 의한 As/P 교환반응이 InAs 양자점의 형상 및 광학적 특성에 미치는 영향을 연구하였다. 시료는 수직형 저압 Metal Organic Chemical Vapor Deposition (MOCVD)를 이용하여 $520^{\circ}C$의 온도에서 성장하였다. 그림1(a) 구조의 양자점은 InP (100) 기판위에 InP buffer layer를 성장한 후 InP와 격자상수가 일치하는 $1.1{\mu}m$ 파장의 InGaAsP barrier를 50 nm 성장하였다. 그 후 As 분위기 하에서 다양한 GI 시간을 주었고 그 위에 InAs 양자점을 성장하였다. 양자점 성장 후 InGaAsP barrier를 50 nm, InP capping layer를 50 nm 성장하였다. AFM측정을 위해 InP capping layer 위에 동일한 GI 조건의 InAs/InGaAsP 양자점을 성장하였고 양자점 성장 후 As분위기 하에 온도를 내려주었다. 그림1(b) 구조의 양자점은 그림1(a) 와 모든 조건은 동일하나 InAs 양자점과 InGaAsP barrier 사이에 GaAs 2ML를 삽입한 구조이다. 양자점 형상 특성 평가는 Atomic force microscopy를 이용하였으며, 광특성 분석은 Photoluminescence를 이용하였다.

• The KIPS Transactions:PartB
• /
• v.18B no.1
• /
• pp.45-50
• /
• 2011

Unknown morpheme errors in Korean morphological analysis are divided into two types: The one is the errors that a morphological analyzer entirely fails to return any morpheme sequences, and the other is the errors that a morphological analyzer returns incorrect combinations of known morphemes. Most previous unknown morpheme estimation techniques have been focused on only the former errors. This paper proposes a unknown morpheme estimation method which can handle both of the unknown morpheme errors. The proposed method detects Eojeols (Korean spacing units) that may include unknown morpheme errors using SVM (Support Vector Machine). Then, using CRFs (Conditional Random Fields), it segments morphemes from the detected Eojeols and annotates the segmented morphemes with new POS tags. In the experiments, the proposed method outperformed the conventional method based on the longest matching of functional words. Based on the experimental results, we knew that the second type errors should be dealt with in order to increase the performance of Korean morphological analysis.

• The Korean Journal of Quaternary Research
• /
• v.14 no.2
• /
• pp.87-99
• /
• 2000

May-June precipitation (253 years : A.D. 1746∼1998) of Yungdong region (coastal area of East-central Korea) was reconstructed using two tree-ring chronologies of Pinus densiflora sampled from Daeseung Fall area in Sorak Mountains. Dry periods were 1765∼1800 (longest dry period), 1835∼1845, 1890∼1910, 1920∼1940 and 1980∼1995, and wet ones 1810∼1830, 1850∼1890 and 1950∼1970. In long-term variation, late 18th century was dry. The 19th century May-June (250㎜) was wetter than the 20th century (231㎜) and the former indicated higher variability than the latter. Major wet/dry periodicities in May-June precipitation series reconstructed were 3 years in short term and 60∼80 years in long term. The present reconstructed data agreed to the ancient rain gauge 'Chukwooki' data (1777∼1907) of Seoul (central-west Korea) in low frequency variations except early 1800s.

• Journal of KIISE:Databases
• /
• v.36 no.2
• /
• pp.84-98
• /
• 2009

Various algorithms and architectures for IP address lookup have been studied to improve forwarding performance in the Internet routers. Previous IP address lookup architecture using Bloom filter requires a separate Bloom filter as well as a separate hash table in each prefix length, and hence it is not efficient in implementation complexity. To reduce the number of hash tables, it applies controlled prefix expansion, but prefix duplication is inevitable in the controlled prefix expansion. Previous parallel multiple-hashing architecture shows very good search performance since it performs parallel search on tables constructed in each prefix length. However, it also has high implementation complexity because of the parallel search structure. In this paper, we propose a new IP address lookup architecture using all-length Bloom filter and all-length multiple hash table, in which various length prefixes are accomodated in a single Bloom filter and a single multiple hash table. Hence the proposed architecture is very good in terms of implementation complexity as well as search performance. Simulation results using actual backbone routing tables which have $15000{\sim}220000$ prefixes show that the proposed architecture requires 1.04-1.17 memory accesses in average for an IP address lookup.