Search

The data for the current research work was collected for 42 different International languages encompassing 3 continents viz. Asia, Europe and South America. The data comprised of unigram model representation of lexicons in the stop-words lists. 13 scripting systems comprising Arabic, Armenian, Bengali, Chinese, Cyrillic, Devanagari, Greek, Gurmukhi, Hanja & Hangul, Kana, Kanji, Marathi, Roman (Latin) and Thai were considered. Based on a comprehensive analysis of statistical measures for Stop-words lists, it has been concluded that Asian languages are mostly self-scripted and that the average number of stop-words in Asian languages is more than those in European languages. In addition to various important and other first research results, a very important inference from the current research work is that the average number of stop-words for any given language could be predicted to be 200.

Many visually impaired people worldwide are unable to travel safely and autonomously because they are physically unable to perceive effective visual information during their daily lives. In this research, we study how to extract the character information of the road sign and transmit it to the visually impaired effectively, so they can understand easier. Experimental method is to apply the Maximally Stable External Region and Stroke Width Transform method in Phase I so that the visually impaired person can recognize the letters on the road signs. It is to convey text information to the disabled. The result of Phase I using samples of simple road signs was to extract the sign information after dividing the exact character area, but the accuracy was not good for the Hangul (Korean characters) information. The initial experimental results in the Phase II succeeded in transmitting the text information on Phase I to the visually impaired. In the future, it will be required to develop a wearable character recognition system that can be attached to the visually impaired. In order to perform this task, we need to develop and verify a miniaturized and wearable character recognition system. In this paper, we examined the method of recognizing road sign characters on the road and presented a possibility that may be applicable to our final development.

Both language experience and processing strategy have been found to affect functional lateralization of the ventral occipitotemporal cortex (vOT). In this study, we adopted a factorial design to investigate the effects of language experience and processing strategy on functional lateralization of different vOT subregions in the processing of familiar (Chinese characters) and unfamiliar characters (Korean Hangul characters) in logographic writings. The processing strategy was manipulated by using part- and whole-based judgement tasks to induce part- and whole-based processing, respectively. The results showed that language experience enhanced neural responses in the anterior and middle vOT subregions, whereas part-based processing enhanced neural activations in the middle and posterior vOT subregions. More importantly, increased neural activations in the left hemisphere induced by language experience and part-based processing resulted in left laterality of the anterior and posterior vOT subregions, respectively, in the processing of logographic characters. These results suggested that functional lateralization of the anterior and posterior vOT subregions were respectively affected by language experience and processing strategy.

The aim of this work is to present a progress of the main research master’s thesis author “Hangeul. The systematic of the writing systems in generating new spatiality”. The writing systems emerged as revolutionary procedures. The societies think them, not only, as original techniques for preservation of traditions, they also give them cultural identity patterns. The Korean writing system called Hangeul is a particular alphabet. It is recognized and registered in the Memory of the World—UNESCO—but is not limited to the field of linguists. Its graphic expression is one of the principal characteristics. Their letters, based only on pure geometry, are drawn quickly and simply. They are designed to be drawn with continuous strokes and based on a square structure. The signs of the Korean alphabet are based on five elements, that is, basic morphological units: points, circles and horizontal, vertical and oblique lines. The experience that is part of the research project Master’s thesis “Hangeul. The systematic of the writing in generating new spatialities” developing by the author that investigate spatial formal explorations into the cutout of the entitative morphology, establishing the systematics of the Korean writing system (Hangeul) as possible starting point of projective processes. The research explores the approaches in analysis of forms cut out of entitative morphology, each sign of Hangeul become the starting point for an initial translation of basic geometric component to design and study the projective logics that establish a conceptual process of translation of the Korean alphabet as source text, as the basis and support of projective processes of forms and spaces.

История корейского протестантизма насчитывает не более полутора веков. За это короткое время он превратился в религию, которая не только укоренилась в Корее, но усилиями теперь уже корейских миссионеров распространяется по всему миру. В статье анализируется начальный этап распространения и адаптации протестантизма в государстве Чосон, открывшегося для внешних влияний в 80-х гг. XIX в. Рассматривается деятельность первых миссионеров, стремившихся действовать в тесном контакте с низшими классами корейского общества. Поразительный успех новой религии во многом объясняется тем, что миссионеры начали свою деятельность с перевода Библии на хангыль – национальный письменный язык, считавшийся образованным классом простонародным. К тому же, исходя из своих установок, они позволяли корейцам трактовать Библию в соответствии с собственными культурными традициями и сами искали параллели с местными религиозными верованиями. Авторы приходят к выводу, что перевод Священного Писания в Корее имел важное историческое и культурное значение: многие корейцы в протестантских церквях впервые приобщились к грамотности, сам хангыль совершенствовался усилиями переводчиков, протестантская трудовая этика стала частью корейского менталитета, через Библию и миссионерские школы происходило знакомство корейцев с демократическими идеями. Все это, помноженное на национализм, способствовало в дальнейшем созданию политической нации, устремленной в будущее., The history of Korean Protestantism is no more than a century and a half. In this short time, it turned into a religion that is not only rooted in Korea, but is now spreading throughout the world through the efforts of Korean missionaries. The article examines the initial stage of dissemination and adaptation of Protestantism in Joseon, opened to external influences in the 80-ies of the XIX century. It deals with activities of the first missionaries, who sought to work in close contact with the lower classes of Korean society. The striking success of the new religion is largely due to the fact that the missionaries started their activities with the translation of the Bible into Hangul, the national written language, which was considered the educated class as a vulgar one. In addition, based on their attitudes, they allowed Koreans to interpret the Bible in accordance with their own cultural traditions and themselves sought parallels with local religious beliefs. The author comes to the conclusion that the translation of the Holy Scriptures in Korea had an important historical and cultural significance: many Koreans in Protestant churches for the first time gained access to literacy; the Hangul was perfected by the efforts of translators; the Protestant labor ethic became part of the Korean mentality; through the Bible and missionary schools, Koreans became acquainted with democratic ideas. All this, multiplied by nationalism, contributed to the further creation of a political nation, looking to the future.

Translating traditional documents is quite laborious and time consuming for human translators owing to the voluminous nature and a complexity of grammatical patterns. In recent times, a neural network-based machine translation architecture such as sequence-to-sequence (seq2seq) model showed superior performance in translation. However, it suffers out-of-vocabulary (OOV) issue when dealing with very complex and vocabulary languages such as Chinese characters, resulting in performance degradation. To cope with the OOV issue, we propose a new method by combining word embedding and character embedding to supplement loss from unknown words with character embedding. Experimental results show that the proposed method is efficient to translate old Korean archives (Hanja) to modern Korean documents (Hangul).

This paper, information communication code standard(KS X 1001, confirmation in 2004), as definded in Hangul Character Code Hangul AMI/HDB-3 the code set for the new system Hangul consonant and vowel tables presented. The result of the existing system and the code set (4x4) bit source coding rules for comparing the frequency of use Hangul consonant and vowel tables(The National Institute of The Korea Language) and statistices showed that 44% of the data processing efficiency is improved.

Thousands of glyphs are necessary for Hangul font generation. It is mandatory to generate the required glyphs before producing Hangul font. This paper, entitled "Multiple Master Glyph Algorithm", presents an process that generates a target number of glyphs automatically from a very small number of glyphs by using a combination rule setting and a glyph interpolation method. A font editor, which is able to generate Hangul glyphs or fonts, is developed based on this algorithm. The editor generates a target number of fundamental glyphs automatically by using a combination rule setting and four master glyphs, which can be set up by a user. The automatically generated glyphs can be used to generate a target font by combining KSX1001 standard Hangul 2350 characters or Unicode standard Hangul 11172 characters automatically. The efficiency of the proposed Hangul editor is analyzed quantitatively in this paper through application to several commercial typefaces.

In accordance with development of many smart phone applications, the importance of keypad that can be used in smart phone has been increasing. In this paper, we design the Hangul smart phone keypad to type a Hangeul characters more efficiently by considering the characteristics of the Hangul characters based on the existing research on smart phones keypad. The proposed keypad in this paper, when we placed the letters on the keyboard, minimizes the travel distance by using the frequency of characters and the associated frequency between vowel and consonant. In addition, we define an assessment model for evaluating the performance of the keypad and verify efficiency of the proposed keypad. According to the result of the experiment, the proposed keypad is more efficient than other keypads.

This study was designed to investigate potential reading differences between low-achieving and typically achieving first-grade Korean-speaking children by (1) comparing their speed of lexical access on alphanumeric and non-alphanumeric tests of rapid automatized naming and (2) comparing relationships between the two groups’ performance on alphanumeric and non-alphanumeric tasks with their performance on component measures of reading. Twenty-seven low-achieving readers and 34 typically achieving readers participated in this study. The low-achieving readers performed significantly more slowly than their typically achieving reading peers on the four measures of rapid naming. The response rates for both alphanumeric naming tasks (digits and letters) showed strong negative correlations with oral reading fluency, while their response rates for the non-alphanumeric naming tasks (Color and Object naming) showed relatively weak relationships with reading measures for both groups. In regression equations, Letter naming tasks were uniquely associated with both word recognition and oral reading fluency. The utility of using alphanumeric naming tasks as indicators for identifying Korean Hangul readers who are at risk for reading disabilities is discussed.

Most mobile devices such as smartphones and tablets have a soft keyboard implemented on the LCD touch screen to receive text inputs from the user. Because the size of the LCD screen is small, many efforts are going on to develop efficient soft keyboard with less key buttons. Especially, Hangul soft keyboard needs much time, effort, and deep software technology in developing an Hangul automata program module. This paper introduces design and implementation details of our authoring tool for Hangul soft keyboard running in Android. The tool supports the user to build an unique soft keyboard easily. The authoring tool downloads the soft keyboard app to the user`s smartphone and makes it run immediately.

Este artículo presenta la evolución histórica de la legua escrita coreana centrándose en el momento histórico de la creación del alfabeto coreano, Hangeul, en el siglo XV. Comienza analizando los orígenes de la lengua coreana y sus conexiones con las lenguas altaicas para luego pasar a ver cómo fue la introducción de los caracteres chinos que fueron el vehículo principal de lenguaje escrito durante la mayor parte de la historia coreana. En los siguientes apartados se explica cómo se creo el alfabeto coreano y sus fundamentos filosóficos; así cómo la oposición que sufrió el Hangeul por parte de algunos miembros de las clases dominantes pero también se convirtió en un elemento esencial para el desarrollo de la cultura entre las clases populares, al mismo tiempo que sirvió de factor unificador en los momentos de crisis nacional.

The last surviving wild population of the Kashmir red deer (Cervus elaphus hanglu) or Hangul inhabits the Dachigam National Park located near Srinagar city of Kashmir, India. This paper presents data from February 2001 to December 2009 on habitat use patterns and food habits of the Kashmir red deer. Investigations based on surveys along trails indicate that Hangul showed significant differences in the use of different habitat types (F = 6.49; P = 0.001) between seasons. Males showed significant differences in habitat utilization both within a season (F = 3.63; P ≤ 0.013) and between seasons (F = 1.50; P ≤ 0.087). Females showed differences in their habitat use within season but not between seasons (F = 14.771; P ≤ 0.577). Surveys showed the deer predominantly using the lower and middle altitudes (1700–1900 m and 1900–2300 m) with no significant gender differences. The riverine and mixed oak forests showed the highest mean deer pellet density/ha (153.71 ± 173.43, n = 418 and 188.60 ± 56.41, n = 114, respectively). The grassy/rocky cliff habitat showed the lowest mean pellet density/ha (55.92 ± 18.32, n = 228). The Hangul diet constituted mainly of dicotyledonous shrubs and trees (45.45 and 80.77%), forbs/herbs (36.36 and 19.23%), monocotyledon grasses and herbs (18.18 and 16.22%) in spring and winter, respectively. Summer diet comprised 42.86% forbs, 19.05% grass/sedges and 38.10% browses. In autumn, Hangul fed mainly on dicotyledonous shrubs. The habitat use patterns of Hangul are comparable with those of the European red deer and are influenced by availability of food, cover, shelter, altitude, slope, aspect and the level of disturbance. The habitats with ideal conditions for utilization but with a meager disturbance were avoided by Hangul. Recommendations are given for better and effective management, protection and conservation of this small and isolated population of red deer.

The purpose of this cross-sectional study was to examine the cognitive abilities that predict reading and spelling performance in Korean children in Grades 1 to 4, depending on expertise and reading experience. As a result, visual cognition, phonological awareness, naming speed and receptive vocabulary significantly predicted reading accuracy in children in Grades 1 and 2, whereas visual cognition, phonological awareness and rapid naming speed did not predict reading accuracy in children in higher grades. For reading, fluency, phonological awareness, rapid naming speed and receptive vocabulary were crucial abilities in children in Grades 1 to 3, whereas phonological awareness was not a significant predictor in children in Grade 4. In spelling, reading ability and receptive vocabulary were the most important abilities for accurate Hangul spelling. The results suggested that the degree of cognitive abilities required for reading and spelling changed depending on expertise and reading experience. Copyright © 2015 John Wiley & Sons, Ltd. Key Messages Phonological awareness and visual cognition were important for Hangul reading. Receptive vocabulary was important in both Hangul reading and spelling. Cognitive abilities are required for literacy changed during reading development.

본 연구의 목적은 유아대상 한글학습용 애플리케이션의 내용을 전통적 언어교육에서 사용하는 언어 교수방법인 발음중심 교수방법과 총체적 교수방법으로 분류하여 비교분석함으로써 학습용 애플리케 이션 내용 개선을 위한 시사점을 찾는 데에 있다. 이를 위해, 선행연구에서 사용된 유아용 애플리케 이션 평가 척도에서 학습내용과 관련된 4개 하위영역을 선택하여 이를 수정, 보완하여 총 51개 애플 리케이션을 대상으로 분석을 실시하였다. 그 결과, 총체적 교수방법의 애플리케이션이 발음중심 교 수방법의 애플리케이션보다 인터랙션을 잘 활용하는 것으로 드러났다. 또한, 총체적 교수방법의 애 플리케이션은 교육의도에 적합한 그래픽의 사용, 매력적인 캐릭터 및 피드백을 통한 능동적 학습활 동 유도에서 발음중심 교수방법의 애플리케이션을 앞섰다. 이 연구결과는 효율적인 한글학습을 위한 양질의 애플리케이션 개발을 위한 기초자료로 활용될 수 있다.

This paper is studied the Korean language to optimize the Hangul character coding for information processing in information terminal device and transmission in network. The paper analyzed Hangul character in Korean language and use frequency of each character. The paper also compared the analysis result to Hangul characters which are coded in standard in Korean character and Unicode. This study referred "Modern Korean Use Frequency Rate Survey Result" issued by The National Institute of the Korean Language. There are total 58,437 Korean words in the report. As a result of this paper, the Korean word 58,437ea are consisted of Hangul character total 1,540ea. The highest use frequency character is "다" and its use frequency to total use frequency rate is 15%. The lowest use character is "휫"and its use frequency to total use frequency rate is 0.00003%. The number of analyzed Hangul character 1,540 is less 7.2 times and 1.5 times than Korean and Unicode standard respectively.

지금까지의 읽기장애 연구에서는 문자언어 음독의 정확성과 속도 즉 해독(음독)능력에 어려움을 가진 난독 증(dyslexia)을 대상으로 한 연구에 초점이 맞추어져 있었으나 최근의 보고에 따르면 음독능력에는 어려움을 가지지 않음에도 불구하고 특수하게 읽기 이해력만이 저하되는 특정 읽기이해부진아동의 사례가 보고되고 있다. 이에 본 연 구에서는 음독능력은 저하되지 않으나 읽기 이해력만이 특수하게 저하되는 읽기이해부진 아동과, 음독능력과 읽기 이해력 모두 저하되는 읽기부진 아동의 특징을 비교검토 하였다. 그 결과 읽기부진아동이 음운인식 과제와 음운 작 업기억 과제 모두에서 유의하게 낮은 수행을 보인 반면 특정 읽기이해부진아동 집단의 경우 음운 작업기억의 유의한 저하만이 나타나는 것을 확인할 수 있었다. 본 연구결과를 토대로 읽기 이해력 장애아동의 특성에 대한 시사점을 논 하였다.

The paper studied possibility of improvement of efficient of data processing in the line coder when Hangul syllables in Unicode system is used for the source code. The scrambling in the line coder is to solve the problem happened due to the source code. The study is based on the HDB-3 scrambling method in ITU-T standards that is applied to AMI line coder. The referred data of Hangul syllables and its use frequency which are required to analysis was used the data extracted from the source data of the National Korean Language Institute. According to the analysis, the average 24% scrambling was generated in source code of Hangul syllables in Unicode system. When the referred Hangul syllables was applied to Unicode system, the average 27% scrambling was producted. Total 8,924ea Hangul syllables in 11,172ea Hangul syllables in Unicode system were not scrambled. Therefore the referred Hangul syllables 1,540ea were accepted in the unscrambled code areas. As a result, the existing Unicode Hangul syllable codes can’t prevent the scrambling, but it is possible to completely remove the 27% scrambling with new source coding system. And then, it can be improved the data processing efficient upto minimum 27% in line coder by software in presentation layer instead of physical layer.

[Purpose] This study investigated the association between proprioception, including joint position sense and kinetic sense, and handwriting legibility in healthy children. [Subjects and Methods] Assessment of joint position sense, kinetic sense, and handwriting legibility was conducted for 19 healthy children. Joint position sense was assessed by asking the children to flex their right elbow between 30° to 110° while blindfolded. The range of elbow movement was analyzed with Compact Measuring System 10 for 3D motion Analysis. Kinetic sense was assessed using the Sensory Integration and Praxis Test. The children were directed to write 30 words from the Korean alphabet, and the legibility of their handwriting was scored for form, alignment, space, size, and shape. To analyze the data, descriptive statistics and Spearman correlation analysis were conducted using IBM SPSS Statistics 20.0. [Results] There was significant negative correlation between handwriting legibility and Kinetic sense. A significant correlation between handwriting legibility and Joint position sense was not found. [Conclusion] This study showed that a higher Kinetic sense was associated with better legibility of handwriting. Further work is needed to determine the association of handwriting legibility and speed with Joint position sense of the elbow, wrist, and fingers.

The aim was to investigate whether native English speakers (Experiment 1) and native Korean speakers (Experiment 2) processed familiar letters in an analytic manner in comparison to unfamiliar letters or symbols. Participants performed a two-alternative-forced-choice identification task with Roman, Korean Hangul and Thai Navon compound letters (large letters made up of small letters). The English speakers were familiar with Roman script but not Korean or Thai, whereas the Korean speakers were familiar with Korean and Roman script but not Thai. The global precedence effect (GPE), an indication of holistic processing, is characterised by a global advantage (global processing is faster than local processing) and asymmetric congruence (global processing interferes with the processing of local features). Based on previous research, it was predicted that there would be a global precedence effect for unfamiliar but not familiar letters. Results from the English speakers did not support this prediction as we found a GPE for familiar Roman as well as unfamiliar Thai letters but not for unfamiliar Korean letters. In contrast, for the Korean speakers, we found support for the prediction as we found a GPE for Thai letters but not for familiar Korean and Roman letters. Based on this evidence, we propose that the Koreans are processing Korean and Roman letters in a more analytic manner than the English native speakers for Roman script. Due to the characteristics of Korean Hangul, Korean readers may be processing letters in a more analytic manner than the English readers.

Background and purpose Korean-speaking patients with a brain injury may show agraphia that differs from that of English-speaking patients due to the unique features of Hangul syllabic writing. Each grapheme in Hangul must be arranged from left to right and/or top to bottom within a square space to form a syllable, which requires greater visuospatial abilities than when writing the letters constituting an alphabetic writing system. Among the Hangul grapheme positions within a syllable, the position of a vowel is important because it determines the writing direction and the whole configuration in Korean syllabic writing. Due to the visuospatial characteristics of the Hangul vowel, individuals with early-onset Alzheimer's disease (EOAD) may experiences differences between the difficulties of writing Hangul vowels and consonants due to prominent visuospatial dysfunctions caused by parietal lesions. Methods Eighteen patients with EOAD and 18 age-and-education-matched healthy adults participated in this study. The participants were requested to listen to and write 30 monosyllabic characters that consisted of an initial consonant, medial vowel, and final consonant with a one-to-one phoneme-to-grapheme correspondence. We measured the writing time for each grapheme, the pause time between writing the initial consonant and the medial vowel (P1), and the pause time between writing the medial vowel and the final consonant (P2). Results All grapheme writing and pause times were significantly longer in the EOAD group than in the controls. P1 was also significantly longer than P2 in the EOAD group. Conclusions Patients with EOAD might require a higher judgment ability and longer processing time for determining the visuospatial grapheme position before writing medial vowels. This finding suggests that a longer pause time before writing medial vowels is an early marker of visuospatial dysfunction in patients with EOAD.

Neurolinguistic circuitry for two different scripts of language, such as phonological scripts (PhonoS) versus logographic scripts (LogoS) (e.g., English versus Chinese, resp.), recruits segregated neural pathways according to orthographic regularity (OrthoR). The purpose of this study was to identify the effect of VSF for cortical representation according to different OrthoR to represent Hangul versus Hanja as PhonoS versus LogoS, respectively. A total of 24 right-handed, native Korean undergraduate students with the first language of PhonoS and the second language of LogoS were divided into high- or low-competent groups for L2 of LogoS. The implicit word reading task was performed using Hanja and Hangul scripts during functional magnetic resonance imaging (fMRI) acquisition. Fluctuations of fMRI BOLD signal demonstrated that the LogoS was associated with the ventral pathway, whereas PhonoS was associated with the dorsal pathway. By interaction analysis, compared with high-competent group, low-competent group showed significantly greater activation for Hanja than for Hangul reading in the right superior parietal lobule area and the left supplementary motor area, which might be due to neural efficiency such as attention and cognition rather than core neurolinguistic neural demand like OrthoR processing.

The purpose of this research is to classify various plants written in the Korean Buddhist scriptures and assign scientific name to them in accordance with the Hangul Tripitaca as standard scriptures, thereby establishing the foundation of the plants in Buddhist scriptures. As a natural resource, we also assessed the value of the plants written in the Buddhist scriptures that distributed from other countries. To this end, we inferred the route of introduction to Korea of the plants and investigated their function and usage. Taxonomic classification of the plants written in Buddhist scriptures identified a total of 331 taxa belonging to 107 families, 244 genera, 313 species, 1 subspecies, 16 varieties, 1 forms. Species composition of the 331 taxa of plants consisted of two taxa of pteridophyte, 15 taxa of gymnosperm, 261 taxa of dicotyledon, and 53 taxa of monocotyledon. Among them, 183 and 148 were woody and herbaceous plants, respectively. Analysis of the distribution of 331 kinds of plants that India's portion of origin of place is 8.9% which holds a low rank compared with those of Korea (10.5%), China (10.9%), and Japan (10.0%). This explains why many taxa of plants distributed from the three countries appeared in the Hangul Tripitaca. Accordingly, we hypothesized that the majority of the plants were added and replaced at some point in past while Buddhist scriptures were propagated from India to China and from China to Korea. Our analysis indicated that 119 out of the 331 taxa (36 %) were distributed from Korea.

In this paper, we examine early efforts in English language education in Korea, focusing on a student’s glossary of Chinese characters that contains Mandarin Chinese, English, and Japanese glosses of the meanings of each character. This book, the Ahak Pyeon, employs a modified version of the Korean Hangul writing system in order to phonetically render the English words, thereby facilitating the acquisition of English pronunciation for Korean learners. We examine the system proposed in this work in detail, noting issues of phonetic accuracy and clarity and assess it with respect to its suitability as a linguistically-sophisticated phonetic transcription system suitable for the representation of English teaching. In this paper we discuss the principles employed in the Ahak Pyeon to render a reasonably accurate phonetic characterization of the pronunciation of the English words included in the book as well as a discussion of the challenges faced by the developer of the system. We discuss this work in terms of the general adaptability of the original Hangul writing system, the creativity of the revised system, and the potential of this system for developing phonemic awareness among Korean learners of English.

In spite of the advances in recognition technology, handwritten Hangul recognition (HHR) remains largely unsolved due to the presence of many confusing characters and excessive cursiveness in Hangul handwritings. Even the best existing recognizers do not lead to satisfactory performance for practical applications and have much lower performance than those developed for Chinese or alphanumeric characters. To improve the performance of HHR, here we developed a new type of recognizers based on deep neural networks (DNNs). DNN has recently shown excellent performance in many pattern recognition and machine learning problems, but have not been attempted for HHR. We built our Hangul recognizers based on deep convolutional neural networks and proposed several novel techniques to improve the performance and training speed of the networks. We systematically evaluated the performance of our recognizers on two public Hangul image databases, SERI95a and PE92. Using our framework, we achieved a recognition rate of 95.96 % on SERI95a and 92.92 % on PE92. Compared with the previous best records of 93.71 % on SERI95a and 87.70 % on PE92, our results yielded improvements of 2.25 and 5.22 %, respectively. These improvements lead to error reduction rates of 35.71 % on SERI95a and 42.44 % on PE92, relative to the previous lowest error rates. Such improvement fills a significant portion of the large gap between practical requirement and the actual performance of Hangul recognizers.