Set Of Chinese Characters Research Articles

Development and Psychometric Evaluation of the Computer-Aided Measure of Chinese Handwriting Legibility (CAM-CHL) for School-Age Children

Handwriting legibility is the main criterion for determining whether a child has handwriting difficulties. A comprehensive assessment of handwriting legibility with sound psychometrics is essential to timely identification of handwriting difficulties and outcome measurement after handwriting interventions. To evaluate the psychometrics of the Computer-Aided Measure of Chinese Handwriting Legibility (CAM-CHL) and to investigate Chinese handwriting legibility in school-age children using the CAM-CHL. Cross-sectional, repeated observation, test-retest. Elementary schools in Taiwan. We recruited 25 lower-grade children for the examination of test-retest reliability, 75 children from all grade levels, and 10 senior schoolteachers for the examination of the CAM-CHL's convergent validity and the investigation of handwriting legibility. Children were asked to copy a set of Chinese characters as legibly as possible. We used the CAM-CHL to assess handwriting legibility in four domains: Size, Orientation, Position, and Deformation. The schoolteachers were asked to subjectively assess the handwriting legibility using a 3-point Likert-type scale. The CAM-CHL demonstrated good to excellent test-retest reliability and acceptable random measurement error in all legibility domains. The CAM-CHL had fair to moderate convergent validity with schoolteachers' perceptions. Additionally, upper-grade children had better handwriting legibility in the Size and Position domains than lower-grade children. The CAM-CHL, a comprehensive and objective method of assessing Chinese handwriting legibility, has sound reliability and acceptable validity, suggesting its potential as an outcome measure for school-age children. What This Article Adds: The CAM-CHL can be used in comprehensive evaluations of Chinese handwriting legibility in school-age children. The CAM-CHL has acceptable psychometrics for use as an outcome measure.

Chinese Calligraphy Generation Based on Involution

The calligraphic works of particular calligraphers often contain only a limited number of characters, rather than the full set of Chinese characters required for typography, which does not meet practical needs. There is therefore a need to develop a complete set of calligraphic characters for calligraphers. Most of the recently popular methods for generating calligraphic characters are based on deep learning, using an end-to-end approach to generate the target image. Deep learning-based methods usually suffer from unsuccessful conversion of stroke structures when the printed font differs significantly from the target font structure. In this paper, we propose an involution-based calligraphic character generation model, which can realize the conversion from printed fonts to target calligraphic fonts. We improve the Pix2Pix model by using a new neural operator, involution, which focuses more on spatial feature processing and can better handle the relationship between strokes than the models using only convolution, so that the generated calligraphic characters have an accurate stroke structure. A self-attentive module and a residual block are also added to increase the depth of the network to improve the feature processing capability of the model. We evaluated our method and some baseline methods using the same dataset, and the experimental results demonstrate that our model is superior in both visual and quantitative evaluation.

On Sino-Turkic verbal functional expressions

AbstractTurkic conquest and rule of China since 386 CE for nearly two hundred years had exerted its profound and long-lasting influence on many levels of Chinese society. Turkic sinification policy had induced the Xianbei National Language (XNL), which was Turkic language with selected set of Chinese characters for phonetic spelling. XNL, being spelt in Chinese characters, managed to function in Turkic-Chinese code-mixing in the bilingual communities as evidenced in bianwen變文. Because of the Turkic politico-socio-economical dominance, some the Turkic elements in the code-mixing eventually gained prominance and have become permanent part of the northern vernacular, predecessor of modern Mandarin. This paper discusses twelve such Turkic-rooted verbal functional expressions: (1). The causative-passivequ取; (2). Transitive passivesha殺,sha煞,si死; (3). Causativedou鬥,dou逗; (4). Continuativehai還,que卻; (5). Resultativeque卻; (6). Reflexivenə呢(7). Positive indirectivemo shi莫是; (8). Negative indirectivebu dao不道; (9). Future participlecai才,cai纔; (10). Conditionalyao shi要是,yao要,yao bu shi要不是; (11). INDUCE-basenong弄; (12). The speech quote verbdao道.

Brain Activity during Visual and Auditory Word Rhyming Tasks in Cantonese-Mandarin-English Trilinguals.

It is unclear whether the brain activity during phonological processing of second languages (L2) is similar to that of the first language (L1) in trilingual individuals, especially when the L1 is logographic, and the L2s are logographic and alphabetic, respectively. To explore this issue, this study examined brain activity during visual and auditory word rhyming tasks in Cantonese–Mandarin–English trilinguals. Thirty Chinese college students whose L1 was Cantonese and L2s were Mandarin and English were recruited. Functional magnetic resonance imaging (fMRI) was conducted while subjects performed visual and auditory word rhyming tasks in three languages (Cantonese, Mandarin, and English). The results revealed that in Cantonese–Mandarin–English trilinguals, whose L1 is logographic and the orthography of their L2 is the same as L1—i.e., Mandarin and Cantonese, which share the same set of Chinese characters—the brain regions for the phonological processing of L2 are different from those of L1; when the orthography of L2 is quite different from L1, i.e., English and Cantonese who belong to different writing systems, the brain regions for the phonological processing of L2 are similar to those of L1. A significant interaction effect was observed between language and modality in bilateral lingual gyri. Regions of interest (ROI) analysis at lingual gyri revealed greater activation of this region when using English than Cantonese and Mandarin in visual tasks.

Neural substrates of bilingual processing in a logographic writing system: An fMRI study in Chinese Cantonese-Mandarin bilinguals

Logographic-logographic bilingualism is lacking in bilingual researches. Cantonese and Mandarin are two Chinese languages that belong to the logographic writing system and share the same set of Chinese characters. Cantonese-Mandarin bilinguals employ logographic-logographic bilingualism, and thus provide a unique population for bilingual studies. In Cantonese-Mandarin bilinguals, it remains unclear whether the second language (L2) phonological processing conforms to the assimilation or accommodation hypothesis and whether the brain mechanisms underlying Cantonese and Mandarin phonological processing are similar. Differential brain mechanisms of Mandarin as first language (L1) and L2 also needs to be clarified. Task-modality functional magnetic resonance imaging (fMRI) was used to explore these questions in Cantonese-Mandarin bilinguals and native Mandarin speakers. Univariate analysis, activation similarity analysis, brain-behavioral analysis and psychophysiological interaction (PPI) analysis were conducted. Novel evidence for accommodation was observed in the L2 processing of Cantonese-Mandarin bilinguals, indicating that small language distance leads to the accommodation of L2 processing, which is absolutely distinct to alphabetic bilingualism. We speculated that the role of language distance in the balance between assimilation and accommodation may be regulated by L1. Notably, we found Cantonese processing relied more on orthography-to-phonology mapping, phonological processing, language control and inhibition while Mandarin relied more on memory extraction. And Mandarin processing as L1 and L2 in our study are highly similar, suggesting the stability of brain regions processing Mandarin.

Correcting Chinese Spelling Errors with Word Lattice Decoding

Chinese spell checkers are more difficult to develop because of two language features: 1) there are no word boundaries, and a character may function as a word or a word morpheme; and 2) the Chinese character set contains more than ten thousand characters. The former makes it difficult for a spell checker to detect spelling errors, and the latter makes it difficult for a spell checker to construct error models. We develop a word lattice decoding model for a Chinese spell checker that addresses these difficulties. The model performs word segmentation and error correction simultaneously, thereby solving the word boundary problem. The model corrects nonword errors as well as real-word errors. In order to better estimate the error distribution of large character sets for error models, we also propose a methodology to extract spelling error samples automatically from the Google web 1T corpus. Due to the large quantity of data in the Google web 1T corpus, many spelling error samples can be extracted, better reflecting spelling error distributions in the real world. Finally, in order to improve the spell checker for real applications, we produce n-best suggestions for spelling error corrections. We test our proposed approach with the Bakeoff 2013 CSC Datasets; the results show that the proposed methods with the error model significantly outperform the performance of Chinese spell checkers that do not use error models.

A Hybrid Ranking Approach to Chinese Spelling Check

We propose a novel framework for Chinese Spelling Check (CSC), which is an automatic algorithm to detect and correct Chinese spelling errors. Our framework contains two key components: candidate generation and candidate ranking . Our framework differs from previous research, such as Statistical Machine Translation (SMT) based model or Language Model (LM) based model, in that we use both SMT and LM models as components of our framework for generating the correction candidates, in order to obtain maximum recall; to improve the precision, we further employ a Support Vector Machines (SVM) classifier to rank the candidates generated by the SMT and the LM. Experiments show that our framework outperforms other systems, which adopted the same or similar resources as ours in the SIGHAN 7 shared task; even comparing with the state-of-the-art systems, which used more resources, such as a considerable large dictionary, an idiom dictionary and other semantic information, our framework still obtains competitive results. Furthermore, to address the resource scarceness problem for training the SMT model, we generate around 2 million artificial training sentences using the Chinese character confusion sets, which include a set of Chinese characters with similar shapes and similar pronunciations, provided by the SIGHAN 7 shared task.

A Critical Edition of Ezra Pound’s Pisan Cantos: Problems and Solutions

If for no other reason the fact that all published editions of the Pisan Cantos have been unable to execute Ezra Pound’s instructions for the insertion of Greek and have omitted over fifty sets of Chinese characters that he directed his publishers to include would be sufficient cause to re-edit the poem. But the case for a new edition is stronger than that. Owing to the extraordinary conditions of its composition and transmission, approximately five hundred corruptions of Pound’s typescript text survived into the poem’s first English and American publications. Pound’s typescript of the Pisan Cantos was prepared during his incarceration in the US Army Disciplinary Training Center (DTC) an American military prison camp near Pisa during the summer and autumn of 1945, and was the product of the harsh conditions of his imprisonment (his wavering memory following a mental breakdown, his lack of books, and the many errors he inevitably produced typing at odd hours on unfamiliar typewriters). Just as seriously, many of Pound’s emended carbons never reached his editors at New Directions and he was forced during his continued incarceration in the U.S. to delegate responsibility for many kinds of correction he would normally have made himself. He was also denied access to his originals at every stage proofing. Although there can never be a definitively “corrected edition” of the Pisan Cantos because Pound made inconsistent emendations on different typescript leaves and kept his publishers in the dark about which instances of idiosyncratic spelling, quotation, and punctuation he wished them to correct, this does not mean that an edition cannot be established that eliminates the corruptions that later crept into the text and that carries out Pound’s implicit and explicit expectations for producing the poem, many of which his first editors never fully understood. A critical edition based on the typescripts Pound produced at Pisa and including a complete historical apparatus is currently in preparation with Oxford University Press and will not only achieve these aims but also make it possible to understand the deficiencies of the poem’s currently circulating texts.

Data-driven Enhancement of Chinese Calligraphy Aesthetic Style

The generating of a large-scale Chinese character set from a small one has been notoriously challenging due to the complexity of character topology and the inherently elusive sophistication in glyphic aesthetics. This paper proposes an innovative approach to synthesize character-glyph, via which a large-scale character glyph set was generated modeling on a small-scale one of a desired style. The approach initiated from the sampling of designated calligraphic works and proceeded to build a character stroke database, followed by the proposal of F-histogram-based character topology. Drawing on aesthetic intuition, we abstract glyphic aesthetics to establish several evaluative rules, and we further designed an algorithm for the evaluation of the character topology with the Support Vector Machine (SVM) algorithm. At last, we adopted simulated annealing algorithm to optimize character glyphs with the desired style(s). Comparatively, this approach deserves credit in that the representation via F-histogram character topology accommodates more types of character topology and the synthesizing of glyphs integrates the stroke shape and the character topology.

Modeling acuity for optotypes varying in complexity

Watson and Ahumada (2008) described a template model of visual acuity based on an ideal-observer limited by optical filtering, neural filtering, and noise. They computed predictions for selected optotypes and optical aberrations. Here we compare this model's predictions to acuity data for six human observers, each viewing seven different optotype sets, consisting of one set of Sloan letters and six sets of Chinese characters, differing in complexity (Zhang, Zhang, Xue, Liu, & Yu, 2007). Since optical aberrations for the six observers were unknown, we constructed 200 model observers using aberrations collected from 200 normal human eyes (Thibos, Hong, Bradley, & Cheng, 2002). For each condition (observer, optotype set, model observer) we estimated the model noise required to match the data. Expressed as efficiency, performance for Chinese characters was 1.4 to 2.7 times lower than for Sloan letters. Efficiency was weakly and inversely related to perimetric complexity of optotype set. We also compared confusion matrices for human and model observers. Correlations for off-diagonal elements ranged from 0.5 to 0.8 for different sets, and the average correlation for the template model was superior to a geometrical moment model with a comparable number of parameters (Liu, Klein, Xue, Zhang, & Yu, 2009). The template model performed well overall. Estimated psychometric function slopes matched the data, and noise estimates agreed roughly with those obtained independently from contrast sensitivity to Gabor targets. For optotypes of low complexity, the model accurately predicted relative performance. This suggests the model may be used to compare acuities measured with different sets of simple optotypes.

Lexical and sublexical semantic preview benefits in Chinese reading.

Semantic processing from parafoveal words is an elusive phenomenon in alphabetic languages, but it has been demonstrated only for a restricted set of noncompound Chinese characters. Using the gaze-contingent boundary paradigm, this experiment examined whether parafoveal lexical and sublexical semantic information was extracted from compound preview characters. Results generalized parafoveal semantic processing to this representative set of Chinese characters and extended the parafoveal processing to radical (sublexical) level semantic information extraction. Implications for notions of parafoveal information extraction during Chinese reading are discussed.

Similarity and Degree of Perplexity Analysis of Chinese Characters

The study of Chinese scripts has always been a topic drawing continuous scholarly attention from various perspectives. Facilitated with the development of computing technology, recent years saw an increasing interest in the graphic pattern of Chinese characters. This paper focuses on the likelihood of orthographical confusions among any given set of Chinese characters. By defining a normalized metric based on character stroke sequences, we introduce the similarity between two Chinese characters, a unique numerical value which falls within the interval [0, 1], measuring to what extent one character is prone to be confused with another. For a set consisting of a large quantity of characters, we introduce the concept of degree of perplexity (DP), measuring the number of strokes weighted average similarities between a given character and the rest of the characters in the set. An efficient and easy-to-implement algorithm is designed to compute the similarity and degree of perplexity. Our formulas are calibrated with numerical experiments simulated with the most frequently used 200 characters. Based on the numerical simulations, an exponential functional relationship between the degree of perplexity and the number of strokes is proposed and is calibrated with least square regression. Finally, possible applications of the measures introduced are discussed.

初等學校『漢字』 敎育을 위한 基本語彙·基礎漢字 選定의 方法 硏究

初等學校『漢字』 敎育을 위한 基本語彙·基礎漢字 選定의 方法 硏究

스마트폰 상에서의 개인화 학습을 위한 랭크부스트 알고리즘

In this paper, we propose a personalized Chinese character education system using RankBoost algorithm on a smartphone. In a typical Chinese character education scenario, a trainee is supplied with a finite number of Chinese characters as an input set in the beginning. And, as the training session repeats, the trainee will notice her/his difficult characters in the set which she/he hardly answers. Those characters reflect their personalized degrees of difficulty. Our proposed system constructs these personalized degrees of difficulty using RankBoost algorithm. In the beginning, the algorithm start with the set of Chinese characters, of which each is associated with the same weight values. As the training sessions are repeated, the algorithm increase the weights of Chinese characters that the trainee mistakes, thereby eventually constructs the personalized difficulty degrees of Chinese characters. The proposed algorithm maximizes the educational effects by having the trainee exposed to difficult characters more than easy ones.

Co‐word analysis using the Chinese character set

AbstractUntil recently, Chinese texts could not be studied using co‐word analysis because the words are not separated by spaces in Chinese (and Japanese). A word can be composed of one or more characters. The online availability of programs that separate Chinese texts makes it possible to analyze them using semantic maps. Chinese characters contain not only information but also meaning. This may enhance the readability of semantic maps. In this study, we analyze 58 words which occur 10 or more times in the 1,652 journal titles of the China Scientific and Technical Papers and Citations Database. The word‐occurrence matrix is visualized and factor‐analyzed.

Pattern recognition using asymmetric attractor neural networks

The asymmetric attractor neural networks designed by the Monte Carlo- (MC-) adaptation rule are shown to be promising candidates for pattern recognition. In such a neural network with relatively low symmetry, when the members of a set of template patterns are stored as fixed-point attractors, their attraction basins are shown to be isolated islands embedded in a "chaotic sea." The sizes of these islands can be controlled by a single parameter. We show that these properties can be used for effective pattern recognition and rejection. In our method, the pattern to be identified is attracted to a template pattern or a chaotic attractor. If the difference between the pattern to be identified and the template pattern is smaller than a predescribed threshold, the pattern is attracted to the template pattern automatically and thus is identified as belonging to this template pattern. Otherwise, it wanders in a chaotic attractor for ever and thus is rejected as an unknown pattern. The maximum sizes of these islands allowed by this kind of neural networks are determined by a modified MC-adaptation rule which are shown to be able to dramatically enlarge the sizes of the islands. We illustrate the use of our method for pattern recognition and rejection with an example of recognizing a set of Chinese characters.

User adaptive handwriting recognition by self-growing probabilistic decision-based neural networks

It is generally agreed that, for a given handwriting recognition task, a user dependent system usually outperforms a user independent system, as long as a sufficient amount of training data is available. When the amount of user training data is limited, however, such a performance gain is not guaranteed. One way to improve the performance is to make use of existing knowledge, contained in a rich multiuser data base, so that a minimum amount of training data is sufficient to initialize a model for the new user.We mainly address the user adaption issues for a handwriting recognition system. Based on self-growing probabilistic decision-based neural networks (SPDNNs), user adaptation of the parameters of SPDNN is formulated as incremental reinforced and antireinforced learning procedures, which are easily integrated into the batched training procedures of the SPDNN. In this study, we developed 1) an SPDNN based handwriting recognition system; 2) a two-stage recognition structure; and 3) a three-phase training methodology for a) a global coarse classifier (stage 1); b) a user independent hand written character recognizer (stage 2); and c) a user adaptation module on a personal computer. With training and testing on a 600-word commonly used Chinese character set, the recognition results indicate that the user adaptation module significantly improved the recognition accuracy. The average recognition rate increased from 44.2% to 82.4% in five adapting cycles, and the performance could finally increase up to 90.2% in ten adapting cycles.

Peripheral and global features for use in coarse classification of Chinese characters

In this paper, a simple and effective approach to the coarse classification of handwritten Chinese characters is proposed. In our approach, a Chinese character is characterized by string representation using periphery and global feature vectors. The peripheral features include four strings to represent the structure of segments in top, bottom, left, and right directions. The global features include the number of horizontal segments in the top direction and bottom direction, and the number of stroke segments in a character. In addition, a scoring-based coarse classification scheme is devised in choosing the proper candidate characters. Twenty sets of Chinese characters (5401 characters/set) are tested. The number of candidate characters is reduced from 5401 to about 80 with the error rate less than 1.2% in average. Experimental results reveal the feasibility of the proposed approach in classifying Chinese characters.

Assessment of visual memory: a comparison of Chinese characters and geometric figures as stimulus materials.

The study compared the suitability of geometric figures and Chinese characters for assessing visual memory. In Study 1, 40 Chinese characters were found to be significantly less verbalisable than were 40 geometric figures from the Biber Figure Learning Test. In Study 2, memory for Chinese characters, for Biber figures, and for items on the Rey Auditory Verbal Learning Test were compared using 14 subjects with left cerebrovascular accident, 15 subjects with right cerebrovascular accident, and 29 matched controls. Subjects in the left hemisphere group showed impairment on the Rey Auditory Verbal Learning Test but not on Chinese characters, whereas subjects in the right hemisphere group showed impairment on Chinese characters but not on the Rey Auditory Verbal Learning Test. This double dissociation was not evident when comparisons involved the Biber Figure Learning Test. The Chinese character set used in this study was thus judged to be more suitable than were geometric figures in the Biber Figure Learning Test for assessing visual memory.

RESEARCH ON CHINESE OCR IN TAIWAN

This paper describes typical research on Chinese optical character recognition in Taiwan. Chinese characters can be represented by a set of basic line segments called strokes. Several approaches to the recognition of handwritten Chinese characters by stroke analysis are described here. A typical optical character recognition (OCR) system consists of four main parts: image preprocessing, feature extraction, radical extraction and matching. Image preprocessing is used to provide the suitable format for data processing. Feature extraction is used to extract stable features from the Chinese character. Radical extraction is used to decompose the Chinese character into radicals. Finally, matching is used to recognize the Chinese character. The reasons for using strokes as the features for Chinese character recognition are the following. First, all Chinese characters can be represented by a combination of strokes. Second, the algorithms developed under the concept of strokes do not have to be modified when the number of characters increases. Therefore, the algorithms described in this paper are suitable for recognizing large sets of Chinese characters.