Search

With the enormous advance of communication technology, telephone numbers to be memorized is increased because most people have a lot of telephone numbers. Thus it needs a method for people to easily memorize those telephone numbers. A traditional method is to use Korean nickname by 4 syllable Korean words such as "Pal -li Pal-li," "Yi- sa Yi-sa" whose initial sounds are corresponding to Korean sounds of telephone numbers. However, this method does not apply to all the telephone numbers, and only apply to a few telephone numbers. This paper proposed a Hangul-phone keyboard (TELNICK) considering the reduction of finger`s moving distance and the number of strokes, and the generation of Korean nicknames for easy memorizing. This paper also proposes a method to generate many Korean nicknames that can correspond to 4-digit or 3-digit telephone numbers by using TELNICK keyboard and presents the form of Korean nicknames that can correspond to 7-digit or 8-digit telephone numbers.t telephone numbers.

Two experiments investigated whether the phonological properties of visually presented words influence the process of word identification. Experiments 1(a) and 2 examined masked phonological priming effects when naming was the response measure. Two types of effect were demonstrated, a so-called ''onset effect'' and homophone priming effect. Experiment 1(a) found an onset effect for primes and targets written in alphabetic Hangul. This effect was larger the more initial letters were shared but there was no effect when the prime and target shared only final letters. These results suggest that the effect was generated in a serial fashion. A companion lexical decision version Experiment 1(b) showed no onset effects. Experiment 2 capitalised on a feature of the Korean writing system whereby homophonic primes and targets can have no orthographic overlap. In this experiment, homophone priming was found in the absence of any onset effects. It was concluded that the phonology of a printed word is processed rapidly...

In this paper, we realized a system that converts the Korean alphabet (Hangul) images, which are in any domain that is formed by the finger movement on the Hangul document, to the editable characters and then outputs them to the word editor. The domain of hand is separated from the sphere of document in the pre-process step of image. The centroid point of hand is drawn by the maximum circular movement method. After the system recognizes the hand with the circular pattern vector algorithm, finds out the position of finger by the distance spectrum and then draws out the sphere of selected character image by the finger movement to divide the characters into character units by applying the histogram between the Hangul characters. We standardized the characters of various sizes. We used the circular pattern vector algorithm that grafts on the fuzzy inference to divert the character images of the domain, which user wants, to the editable characters by comparing the characteristic vectors between the standard pattern character and the inputted character and by recognizing the character.

In this paper, we propose a grapheme verifier based on a neural network (NN) for on-line handwritten Hangul (Korean script) recognition. The verifier complements the global information and discrimination power to hidden Markov models (HMMs) by penalizing incorrect inputs and passing correct ones intact. It takes structural features from partial inputs aligned to HMM states as global information and gives the probability of their validity; one if they belong to the grapheme and zero if they do not. Hangul syllable characters are recognized with the network of grapheme models which are composed of HMM recognizers and NN verifiers. Probabilities of verifiers are integrated with those of HMMs during the search for the most probable path in the network. Experimental result showed that the proposed verifiers effectively reduced recognition errors of HMM recognizers for Hangul syllable characters by 39.2% on average.

In this paper, the author proposes new canonical decomposition (NFD: normalization form D) and composition processes (NFC: normalization form C) for Hangeul, Korean script. The new processes can handle both Modern and Old Hangeul in a consistent way. Therefore, the new processes are more general than the ones described in UAX #15, which handle only Modern Hangeul syllables.Since NFD and NFC will play an important role in sorting and searching, Internet-related applications such as multilingual domain names, databases, etc., we need to establish good NFD and NFC for Hangeul before they are widely used.

Diphthongs have formant transitions sensitive to speaking rate, stress, vowel quality, and language ability. Tongue movement typically shows a functional “pivot” where the palate to tongue distance is almost constant; some combinations result in an “arch,” with only part of the tongue moving [Iskarous, J. Phon. 33, 363-381, 2005]. Here, ultrasound images of the tongue in Korean, Japanese, and Mandarin diphthongs and triphthongs (Mandarin) (vowel sequences in Korean and Japanese) were analyzed. Five repetitions of 6 nonwords were presented in Hangul, Hiragana or both Pinyin and a character for a word with the same vowel on tone 1. Diphthongs were [ai, ei, au] and triphthongs/sequence of three vowels were [iau, ieu, and uai]. All the ultrasound image frames for the target vowels were traced and superimposed. Preliminary results indicate that most vowel quality pairs resulted in a pivot pattern, with some arch patterns as well. Even in Mandarin triphthongs, there were generally two pivots, even though phonologically the sequence is considered a single vowel nucleus. Several approaches to quantifying this effect will be presented. It is possible that the success in producing a pivot could indicate mastery of production, both in development and in second language learning.

For patient information identification, demographic fields (name and address), and medical information (medical history) are compared in order to measure the similarities. The improvement of algorithm related to Koreans’ Hangul name is necessary in the individual distinguishing part in order to introduce MPI system matching Korea. Since Hangul is a type of combination, we cannot match the names more exactly by the one-dimensional distance comparative function which was used in existing English. This paper suggests the algorithm for the Korean name comparison.

© 2015 Elsevier Inc. Previous studies have suggested differential engagement of addressed and assembled phonologies in reading Chinese and alphabetic languages (e.g., English) and the modulatory role of native language in learning to read a second language. However, it is not clear whether native language experience shapes the neural mechanisms of addressed and assembled phonologies. To address this question, we trained native Chinese and native English speakers to read the same artificial language (based on Korean Hangul) either through addressed (i.e., whole-word mapping) or assembled (i.e., grapheme-to-phoneme mapping) phonology. We found that, for both native Chinese and native English speakers, addressed phonology relied on the regions in the ventral pathway, whereas assembled phonology depended on the regions in the dorsal pathway. More importantly, we found that the neural mechanisms of addressed and assembled phonologies were shaped by native language experience. Specifically, one key region for addressed phonology (i.e., the left middle temporal gyrus) showed greater activation for addressed phonology in native Chinese speakers, while one key region for assembled phonology (i.e., the left supramarginal gyrus) showed more activation for assembled phonology in native English speakers. These results provide direct neuroimaging evidence for the effect of native language experience on the neural mechanisms of phonological access in a new language and support the assimilation-accommodation hypothesis.

We compare the cognitive processing of words written in alphabetic scripts with the cognitive processing of words written in logographic scripts. We suggest that the processing of words written in alphabetic scripts relies more heavily on the storage of--and the serial rehearsal properties of--short-term memory's phonological loop. In contrast, the processing of words written in logographic scripts relies more on the storage of--and the spatial-relational rehearsal properties of--visual short-term memory. A series of three experiments investigates implications of these processing differences within a single language, Korean, where words can be written in the alphabetic Hangul or in the logographic Hancha. These experiments examine contextual interference from auditory and visual stimuli, relational memory between brand names and auditory and visual brand identifiers, and two qualitative processing outcomes, serial-order memory and spatial-relational memory. Copyright 2001 by the University of Chicago.

In the present studies, we examined strategic flexibility in the use of lexical and sublexical information in Korean word recognition. Korean readers show a large frequency effect for words printed in the alphabetic Hangul script only if these words are embedded in a list consisting largely of logographic (Hanza) words. In the first experiment, high- and low- frequency Hangul words were preceded by Hangul or Hanza words. Frequency effects were eliminated when a Hangul target word was preceded by two Hangul words, even when the overall proportion of Hanza words in the list was large. In the second experiment, one group saw each stimulus preceded by a cue indicating which script would be seen on that trial. Frequency effects for Hangul targets were absent under this condition, but were present when subjects were not so cued. These results indicate that Korean readers are able to control their use of lexical and sublexical information over a small number of stimuli or even trial by trial.

A hierarchical random graph (HRG) representation for handwritten character modeling is presented. Based on the HRG, a Hangul, Korean scripts, recognition system also has been developed. In the HRG, the bottom layer is constructed with extended random graphs to describe various strokes, while the next upper layers are constructed with random graphs (Wong and Ghahraman, IEEE Trans. Pattern Anal. Mach. Intell. 2(4) (1980) 341) to model spatial and structural relationships between strokes and between sub-characters. As the proposed HRG is a stochastic model, the recognition is formulated into the problem that chooses a model producing maximum probability given an input data. In this context, a matching score is acquired not by any heuristic similarity function, but by a probabilistic measure. The recognition process starts from converting an input character image into an attributed graph through the preprocessing and the graph representation. Matching between an attributed graph and the hierarchical graph model is performed bottom-up. Since the hierarchical structure in an attributed graph is decided after the recognition ends depending on the best interpretation of the graph matching, we can avoid incorrect sub-character segmentation. Model parameters of the hierarchical graph have been estimated automatically from the training data by EM algorithm (Dempster et al., J. Roy. Stat. Soc. 39 (1977) 1) and embedded training. The recognition experiments conducted with unconstrained handwritten Hangul characters show the usefulness and the effectiveness of the proposed HRG.

This paper was presented under the title "How has the Korean vowel system changed through time?" at the 11 th International Conference on Korean Linguistics (Honolulu, July 1998). There are three parts: (1) on Korean vowel harmony; (2) on the hypothesis that the values of the modem vowel system is the result of rotation, a shift of articulatory values; (3) on the connotative associations of the vocalic features, the etymological implications of vowel sets, and the canonical patterns found in the nouns and the verb stems.

The Korean orthography includes alphabetic Hangul and logographic Hanja scripts. Two experiments investigated whether (a) the role of phonological and orthographic activation differs during semanti...

Orthographies vary in the support they provide for word identification based on grapheme-phoneme correspondences. If skills developed in acquisition of first-language (L1) reading transfer to reading English as a foreign language (EFL), the extent to which EFL readers' word identification shows reliance on information other than grapheme-phoneme correspondences could be expected to vary with whether their L1 orthography is a non-Roman alphabet such as Korean hangul or a nonalphabetic (morpho-syllabic) system such as Chinese characters. Another influence could be whether EFL readers have learned to read a morpho-syllabic L1 by means of an alphabetic transliteration. English text reading speeds and oral reading quality ratings of three groups of adult Asian EFL readers attending an American university were compared with those of two groups of American L1 readers: Graduate student peers and eighth-grade students. All EFL groups read more slowly than both groups of L1 readers, and their reading was more impaired when orthographic cues were disrupted by mixed case print or pseudohomophone spellings. Some of these effects were reduced in EFL readers from Hong Kong, who had earlier exposure to English. Contrary to previous findings, no effects could be attributed to type of first orthography or early exposure to alphabetic transliteration of Chinese characters, which differentiated the Taiwanese and Hong Kong groups. As a whole, the results suggest that, at least across the L1 groups studied, differences in EFL word reading are associated less with type of L1 orthography than with history of exposure to English.

In this paper, we propose a method of character extraction from documents in which both Hangul (Korean characters) and alphanumeric characters are written. In order to ensure accurate segmentation of touching characters, character segmentation and recognition are performed by turns. We use a recognizer to select the correct cutting point among candidates generated by the character segmenter. The character segmenter is implemented by a Multi-Layer Perceptron (MLP) trained by a back propagation algorithm. As the MLP has been trained for all kinds of touching characters, it can segment all touching characters by itself. Experimental results show that the proposed method achieves high segmentation rate in documents written in both Hangul and alphanumeric characters.

This paper presents (1) a quantitative performance comparison of the features used in off-line handwritten Korean alphabet recognition, and (2) some experimental results based on the feature combination approach. We classified the features into three: geometrical/topological, statistical and global. For each class, we selected four or five features and performed recognition experiments with PE92 database using neural network classifiers. We compared their recognition performances and selected the top four among those thirteen features: Up Down Left Right Hole, Gradient, Structure, and Mesh. With the combination of some or all of the four features, we repeated recognition experiments. Experimental results show that (1) as a whole, geometrical/topological features outperform the other two feature classes in terms of the recognition rate, and (2) UDLRH and Gradient features in geometrical/topological feature class outperform the other features, (3) the feature combination approach can result in performance improvement.

This paper includes a description of 3 affiliated oriental languages: Chinese, Japanese, and Korean. It includes a description of the origins of these 3 languages and the inter-relationship among them. Drawn from the viewpoints of several experienced researchers in the field of OCR (Optical Character Recognition) and computational linguistics, it attempts to bring out the intriguing aspects of these 3 ideographic languages, including the formation and composition of pictograms, special features, learning, understanding, contextual information, and recognition of characters and words, and their relations to poetic expressions and pattern recognition techniques. Numerous references are given and comments on future trends are also presented.

There are two types of composition systems for the Korean alphabet: a combined system and a composite system. This paper describes an efficient coding method for both of these two types. Using this coding method, with the combined system yields about 10.5% code-length savings per a Korean character, while it yields about 45% savings with the composite system. In other words, the coding method produces a better result (i.e. 34.5% better) with the composite system than with the combined system. The simulation has been performed on a MasPar machine having 64 processors. The results show that the combined system achieved a 45.851-fold speedup while the composite system achieved a 47.274-fold speedup.

Individuals with grapheme-color synesthesia experience “colors” when viewing achromatic letters and digits. Despite the large individual difference in synesthetic association between inducing graphemes and induced colors, the search for the determinants of synesthetic experience has begun. So far, however, research has drawn an inconsistent picture; some studies have shown that graphemes of similar visual shape tend to induce similar synesthetic colors, while others suggested sound as an important factor. Moreover, meaning seems to affect synesthetic color. In the present work, we sought to investigate the determinants of synesthetic color by testing four multilingual grapheme-color synesthetes who experience “colors” upon viewing Korean (hangul), Japanese (katakana and hiragana), and English (Latin alphabet) characters on a standardized color-matching procedure. Results showed that pairs of characters of matched sound tended to induce similar synesthetic colors. This was the case not only between two scripts within the same language (Japanese hiragana and katakana) but also between two different languages (Japanese and Korean). In addition, pairs of characters with similar initial phonemes tended to induce similar colors; this was general across multiple languages. Results also showed that pairs of sequential words in Korean, Japanese, English, and Chinese that have the same meaning tended to elicit similar synesthetic colors. When those pairs of words shared not only meaning but also sound, the similarity of the induced synesthetic colors was even greater. Our work is one of the few initial attempts to examine the influence of visual shape, sound, meaning, and their interaction on synesthetic color induced by characters across multiple languages.

The cover of a book of Kim Dae-jung’s speeches and interviews, Philosophy & Dialogues: Building Peace & Democracy, published in 1987 in Hangul and English, proudly stated that in January 1987 DJ “was honored by being nominated for the Nobel Peace Prize in recognition of his three-decade-long struggle for democracy and human rights.” The cover blurb said the nomination was “a tribute to Kim Dae-jung and an honor for the Korean people he has so proudly served.”1 The blurb might also have added that DJ was the hero of the American intellectual establishment, a figure comparable to South Africa’s Nelson Mandela, as popular in the groves of American academe as he was on the campaign trail in his native Cholla region.

This paper describes an efficient coding method for Korean characters (alphabet) using a three-state transition graph. Parallel Hangul Dynamic Coding Method (PHDCM) compresses about 3.5 bits per Korean character compared with other coding techinques. When we ran the method on a MasPar machine, it achieved a 49.314-fold speedup with 64 processors having 10 million Korean characters

This paper proposes a recognition system of constrained Handwritten Hangul (Korean character) and alphanumeric characters using discrete hidden Markov models (HMMs). The HMM process encodes the distortion and similarity among patterns of a class through a doubly stochastic approach. Characterizing the statistical properties of characters using selected features, a recognition system can be implemented by absorbing possible variations in the form. Hangul shapes are classified into six types by fuzzy according to their effectiveness in each class. The constrained alphanumerics recognition is also performed using the same features employed in Hangul recogndition. The forward-backward, Viterbi and Baum-Welch reestimation algorithms are used for training and recognition of handwritten Hangul and alphanumeric characters. The simulation result shows that the proposed method recognizes effectively handwritten Korean charaters and alphanumerics

There are many uncertainties in handwritten character recognition. Stochastic modeling is a flexible and general method for modeling such problems and entails the use of probabilistic models to deal with uncertain or incomplete information. This paper presents an efficient scheme for off-line recognition of large-set handwritten characters in the framework of stochastic models, the first-order hidden Markov models (HMMs). To facilitate the processing of unconnected patterns and patterns with isolated noises, four types of feature vectors based on the regional projection contour transformation (RPCT) are employed. The recognition system consists of two phases. For each character, in the training phase, multiple HMMs corresponding to different feature types of RPCT are built. In the classification phase, the results of individual classifiers to produce the final recognition result for an input character are integrated, where each individual HMM classifier produces one score that is the probability of generating the test observation sequence for each character model. In this paper, several methods for integrating the results of different classifiers are considered so that a better result could be obtained. In order to verify the effectiveness of the proposed scheme, the most frequently used 520 types of Hangul characters in Korea have been considered in the experiments. Experimental results indicate that the proposed scheme is very promising for the recognition of large-set handwritten characters with numerous variations.

Chinese characters and Buddhist books reached Korea about third century of the Common Era. From that moment Korea became the official part of the Chinese culture and civilization (Cheon So-yeong 2007, 362). Despite the fact that ideographic Chinese characters were not fit for the Korean language, Koreans, having no writing system of their own, adopted the script. Chinese characters became the basis for the Korean writings and official documents till mid-19 th century. Initially, texts were formulated in Chinese. The sentence order typical of the Chinese language was used in them too. At the same time Koreans read out such texts adapting the pronunciation of Chinese characters to the rules of Korean phonetics. With the flow of time, new styles of writing started emerging including gugyeol and idu. Gugyeol was a style in which the sentence word order was still typical of the Chinese language. It was, however, enriched with Korean grammatical morphemes which were absent in Chinese but were vital for Korean. The next stage in the development of Korean writings is called idu. What is typical of idu is the fact that the sentence word order changed into typical of Korean spoken language (e.g. the verb was transferred at the end of the sentence). None of the methods was fully efficient. Therefore, the most significant breakthrough was the invention of the Korean alphabet called hangeul by the king Sejong.

The purpose of this study was to design and develop a natural user interface intended to improve the degree of absorption of users in serious games made for education and propagation of the Korean Language, Hangul. A natural user interface using gesture recognition with general-purpose databases is one method broadly employed for improving the degree of user absorption in games. Many studies have been conducted to find ways of achieving higher levels of user absorption; however, the generic gesture recognition methods designed to be applicable to all types of users cannot be expected to perform favorably in serious games typically equipped with limited objects. The method presented in this study was designed to enable use of characteristic gesture recognition by classifying users into groups of children and foreign adults. The puzzle game used to teach the Korean language was designed specifically to recognize user gestures more efficiently.

This study examined colors in an animation built by the characterization of consonants and vowels of Hangul, the Korean alphabet, with the purpose of increasing the learning effect for infants. It particularly examined the backgrounds in the creation of Hangul, principles used in the invention of Hangul’s consonants and vowels, coloration and its association, color image language, and learning stages according to verbal development. Moreover, the various colorations of the characters and background, differentiated at each stage of a game, were also studied. For the coloration studies, we referred to previous research regarding the coloration of Korean-style colors and the relationship between memory and coloration.

Background Korean orthography is composed of Hanja (ideograms) and Hangul (phonograms). Based on previous studies, the fusiform gyrus has been associated with ideogram reading. We examine serial functional magnetic resonance imaging (fMRI) images in a patient exhibiting dissociation of Hanja and Hangul reading to identify brain areas associated with Hanja reading. Case report fMRI were taken of a 63-year-old man showing profound Hanja alexia with normal Hangul reading after an acute stroke involving the left frontal and parietal lobes, who later spontaneously recovered his Hanja reading ability. Scans were taken while performing Hanja and Hangul reading tasks on three occasions. As a result, in spite of having profound Hanja alexia, partial activation of the fusiform gyrus was observed on the first fMRI. Serial fMRI scans showed activation of the bilateral middle frontal gyri that increased in parallel with the patient's recovery of Hanja reading. Conclusions The frontal lobe, not only fusiform gyrus, may play role in reading Hanja, although more evidence is needed.

This paper presents a system architecture and algorithm for the disambiguation problem in human-robot interaction. Currently, when we have a communication with robot, there are ambiguity problems which lead to a misunderstanding. Conventional methods only identify ambiguity in limited ways and in few contexts due to the cost of doing so. The proposed method using real Hangul input object RHINO cloud identifies ambiguous words, phrases and sentences in many contexts and suggests appropriate alternatives. And by calculating the frequency of an ambiguous word, an associated word and the theme we can obtain the associated strength. The theme which has the biggest strength is the meaning of the ambiguous word. This process reflects the fluctuation of associated words' social cultures because it searches words in real time.

This paper proposes segmentation method in mixed Korean and English printed text which contains touching characters using clustering strategy. At the first step, a vertical projection of image text is determined, and clustering process performed on it. Then the cluster with the smallest mean value used as candidate segmentation point. This process will produce candidate bounding boxes. Furthermore, they should be verified whether according to Korean or English characteristics otherwise they will be splitted or merged each others. The merged process could be done based on Korean vowel characteristics since Korean alphabet consist several symbols, while splitted process could be done by local vertical projection clustering. The proposed method gives 99.36% correct segmentation rate in un-touching characters and 99.25% in touching characters. This result shows that the proposed method using clustering strategy is very effective for touching problem in mixed Korean and English printed text. Besides, it also improves the speed of segmentation process, because the method does not need a character recognizer to verify bounding boxes.

The English alphabet is employed for inputting English and other Western languages into the Computer. It is also used for many other languages such as the Chinese and Japanese for the same purpose. Since the Alphabet is a phonetic system, it can be said that all computer input is made through phonetics. Although the script system of these languages use commonly the Alphabet, the way of reading them differ one another, and one who read a system may not possibly read the others. A question comes to rise. Can't there be a unique universal phonetic system commonly applicable to all languages? The IPA, the most logical candidate, would not be accepted for practical reasons. A simple phonetic script system composed of 10 basic consonants, 10 basic vowels and 10 types of intonations is introduced. The system, based on Hangul, can describe the phonetics of virtually any language through the applications of these basics. The proposition is validated through a mobile application program called HUPS, Hangul-based Universal Phonetic System.

The objective of this study was to develop an edutainment robot which provides multi-sensory learning experiences to improve users' space perception and creativity. In particular, we focused on developing educational content for the study of the Korean alphabet (Hangul) using the robot. On the basis of the phonemic and modular nature of Hangul, we devised a block-shaped edutainment robot for the study of Hangul. The robot known as "HangulBot" is composed of a consonant block and a vowel block. By rotating and rearranging those blocks, a user can create different characters. To enable the robot to perceive the arrangement of the blocks and the distance between a consonant block and a vowel block, IR LEDs and photo transistors were used. The eight IR LEDs in the consonant block generate different radiation signals, and the vowel block perceives the arrangement of the blocks by receiving the signals. The distance between the two blocks is estimated by measuring the thresholding, and the corresponding sound of each arrangement is then played through a speaker installed in the vowel block. We executed two short-term field trials with a twenty-seven month old child in June of 2011 and November of 2011 to ascertain children's initial reaction to HangulBot and how their reaction would change over time. While the results are preliminary, we noted several interesting findings. First, after several trials by the mother, the child felt comfortable with HangulBot. Second, the child intuitively followed the corresponding speech sounds which were generated by HangulBot according to the arrangement of the blocks. That is to say, the sound generated after the arranging the block intuitively induced the child to follow the sound. Third, the child's initial reaction to HangulBot was mostly block play, but after five months later, her reaction to the robot included not only block play but also active learning of the Korean alphabet. This result indicates that HangulBot could be an effective edutainment tool which improves space perception and creativity as well as linguistic abilities by stimulating both sides of the brain.

We propose an interactive toy, "Hangul Gangul", which helps users learn Hangul characters using a tangible interface. Using our system, users can enjoy learning Hangul characters by combining physical blocks representing vowel and consonant characters. Our system aims to encourage collaborative learning between children and adults.

This paper proposes a method to detect the Hangul character region from natural images using the topological structural features of Hangul graphemes. First, I transform a natural image to a gray-scale image. Second, feature extraction is performed with the edge and connected component based method. The edge-based method uses a canny-edge detector and the connected component based method applies local range filtering. Next, if features are not corresponding to the heuristic rule of Hangul characters, extracted features are filtered out and selected candidates of the character region are identified. Next, candidates of the Hangul character region are merged into one Hangul character using a Hangul character merging algorithm. Finally, we detect the final character region by a Hangul character class decision algorithm. The experimental results show that the proposed method could detect character regions effectively in images that contain a complex background and various environments. As a result of the performance evaluation, the proposed method shows advanced results about the detection of Hangul character regions from a mobile image.

Usage of the Korean alphabet, han’gŭl 한글, is for the most part restricted to write Korean itself and so far there has not been any other language it is commonly associated with as its default script. This chapter outlines some of the more important deviations from common usage, largely drawing upon works for the instruction in Chinese, Japanese and Manchu at the Bureau of Interpreters, with some additional materials of later times unrelated to the Bureau. Diacritical marks can be subdivided into two classes: those whose foreign origin is evident at first sight and those for which this does not necessarily hold. The latter category comprises only such diacritics which are so simple in shape - mere dots or short strokes, circles at best - that any claim to a foreign origin suffers from the possibility of independent invention. Keywords:diacritical marks; foreign languages

In this paper, we propose efficient postprocessing algorithms for error correction in handwritten Hangul (Korean script) address and human name recognition. As the load on the character recognizer for the recognition of the administrative district part of addresses was reduced by restricting the number of candidate characters to be matched by using an address lexicon, the processing speed and recognition rate were greatly improved. Also, as the misrecognition results from the character recognizer were corrected by efficient postprocessing algorithms based on backtracking, a high recognition rate of more than 98% could be obtained for the administrative district part of input addresses. For the recognition of human name part, misrecognition could be effectively corrected by combining the a priori probability and the confusion probability of each character which composes human names.

Two experiments assessed word recognition in processes in Korean. The Korean orthography mixes alphabetic and logographic scripts ( Hangul and Hanza , respectively). In the first experiment, native Korean speakers named words printed either in Hangul or in Hanza. One group received 20% Hanza and 80% Hangul, and for the second group these proportions were reversed. Recognition of Hangul words, but not Hanza words, was affected by this manipulation. In the second experiment, subjects read high- and low-frequency Hangul words embedded in a list consisting of additional Hangul words, Hanza words, or Hangul pseudowords. Only in the context of Hanza words did Hangul word frequency affect naming time. These results suggest a flexibility in the use of phonological information in reading alphabetic Korean, and suggest further that a phonological route to word recognition is favored by the shallow Hangul orthography.

Background: Agraphia in Korean patients may be different from agraphia in other patients who use alphabetical writing systems due to the “visuoconstructional script” characteristics of the Korean writing system, Hangul. Patients with early onset Alzheimer's disease (EOAD) have a severe degree of hypometabolism in the parietal area, which is known to be involved in processing visuospatial function. Thus, we explored the diverse error patterns manifested in writing single syllables in Korean patients with EOAD.Methods: A study sample of 35 patients with EOAD and 18 healthy controls (HC) performed a Hangul writing task. We analyzed the erroneous responses of the subjects according to visuoconstructional and linguistic characteristics. In addition, we evaluated the relationship between Hangul writing and the neuropsychological variables as well as the severity of dementia.Results: When comparing the total number of erroneous responses between EOAD and HC groups, the performances of EOAD patients were significantly worse than those of HC. EOAD patients demonstrated visuoconstructional errors even in the early stages of the disease. Severity of dementia and multiple cognitive domains such as attention, language, immediate memory, and frontal executive functions significantly correlated with the performance of Hangul writing.Conclusion: Our findings suggest that patients with EOAD exhibit not only linguistic errors but also visuoconstructional manifestations of agraphia, which are associated with cognitive impairments in the multiple domains.