Word Segmentation Research Articles

Perception of Japanese accented English segments in words

Abstract We studied the perception of segmental boundaries in native English and Japanese accents in terms of foreign accentedness, intelligibility and discrimination accuracy. Five American English vowels /ɪ, æ, ɑ, ə, ɝ/ and seven consonants /r, l, v, θ, f, t, k/ were extracted from English and Japanese words produced by an American English-Japanese bilingual speaker, and manipulated, with the acoustic properties shifting gradually from 100% Japanese to 100% American English using segmental foreign accent and gradation techniques. Perception tests were conducted online with native American English speakers, and groups of Japanese speakers with either low or high English proficiency. All three groups showed a negative correlation between the degree of foreign accent and intelligibility. Also, there was a greater reduction in intelligibility due to acoustic deviation from the native norm in consonants than in vowels. The results suggest that more standard-like accent does not necessarily improve intelligibility, thus highlighting the importance of FA research in terms of segmental accuracy.

Análise de dados de segmentação não convencional de palavras em textos de alunos do Ensino Fundamental I

This article brings some considerations about the non-conventional segmentation of words based on a research study carried out in the first author’s scientific initiation program, in the 2019-2020 cycle. For this research, 51 essays from Elementary School students from two educational institutions in Porto Velho (RO) were collected, quantified and analyzed. After analyzing the data collected based on the theories proposed by current phonological models, such as Prosodic Phonology, a great influence of prosody was perceived in the segmentation elaborations of the writers, soon realizing that by studying this phenomenon it is possible to contribute to the study of the relationship between orality and writing.

A hybrid method of combination probability and machine learning for Chinese geological text segmentation

To address the issues surrounding incomplete coverage of core dictionaries, limited training corpora, and low precision in Chinese geological text segmentation, a knowledge- and data-driven word segmentation method by combining combination probability and machine learning was proposed in this paper. We extracted mathematical feature information from terms in Chinese geological text to construct a Term Combination Probability Model (TCPM) for Chinese word combinations by integrating the combination features of geological terms and the Chinese writing styles under zero-sample conditions. The TCPM was used to extract geological terms with high combination characteristics as a user-defined dictionary, and then a geological corpus was constructed by using a general domain word segmentation method based on this dictionary. After a small amount of manual review and optimization, the geological corpus was trained with a BiLSTM-CRF model to segment Chinese geological text. The proposed method in this paper was tested using a regional geological survey report set in Henan Province, and the precision, recall, and F1-score of the method are 92.65%, 92.53%, and 92.59%, respectively. The experimental results demonstrated that this method, combined with the inherent knowledge features of geological text and the machine learning method, can assist in expanding the core dictionary for Chinese geological text segmentation based on zero-sample, and can improve the segmentation precision of Chinese geological text compared to simple general word segmentation methods or machine learning methods.

Chinese readers utilize emotion information for word segmentation.

We reported a large-scale Internet-based experiment to investigate the impact of emotion information on Chinese word segmentation, in which participants completed an overlapping ambiguous string (OAS) segmentation task and the Chinese version of Beck Depression Inventory-II in a counterbalanced order. OAS is a three-character string (ABC) in which the middle character can form a distinct word with both the character on its left side (word AB) and the character on its right side (word BC). Participants were presented with isolated OASs and were asked to report the word they identified first. Emotional OAS was constructed by a combination of a neutral word and an emotional word, with the neutral and emotional words sharing character B. We orthogonally manipulated the valence of the emotional words (positive vs. negative) and their position in the OAS (left-side vs. right-side). The results showed that compared with neutral words, both positive and negative words were more likely to be segmented, and this segmentation outcome was not affected by readers with different depression tendencies. These findings suggest that emotion information can influence word segmentation, and that both positive and negative words take precedence over neutral words in the word segmentation process. This study provides a new perspective and evidence to understand the impact of emotion information on word processing.

Improved Word Segmentation System for Chinese Criminal Judgment Documents

ABSTRACT In this paper, a system for automatic word segmentation of Chinese criminal judgment documents is proposed. The system uses a hybrid model composed of fine-tuned BERT (Bidirectional Encoder Representations from Transformers), BiLSTM (Bidirectional Long Short Term Memory) and CRF (Conditional Random Field) for named entity recognition, and introduces a custom dictionary that includes common professional terms in Chinese criminal trial documents, as well as a rule system based on judicial system and litigation procedure related regulations, to further improve the accuracy of word segmentation. BERT uses a deep bidirectional Transformer encoder to pre-train general language representations from large-scale unlabeled text corpora. BiLSTM uses two LSTM networks, one for the forward direction and one for the backward direction, to capture the context from both sides of the input sequence. CRF uses a set of features and weights to define a log-linear distribution over the output sequence. Experimental results show that the proposed system has significantly improved word segmentation accuracy compared to the current commonly used Chinese word segmentation models. In the results of the segmentation of the test data, the F1 scores for jieba, THULAC and the segmentation system proposed in this paper are 85.59%, 87.94% and 94.82%, respectively.

Fusion of the word2vec word embedding model and cluster analysis for the communication of music intangible cultural heritage

This article aims to propose a method for computing the similarity between lengthy texts on intangible cultural heritage (ICH), to facilitate the swift and efficient acquisition of knowledge about it by the public and promote the dissemination and preservation of this culture. This proposed method builds on traditional text similarity techniques. The ultimate goal is to group together those lengthy texts on ICH that exhibit a high degree of similarity. First of all, the word2vec model is utilized to construct the feature word vector of music ICH communication. This includes the acquisition of long text data on music ICH, the word segmentation of music ICH communication based on the dictionary method in the field of ICH, and the creation of a word2vec model of music ICH communication. A clustering algorithm analyzes and categorizes ICH communication within the music. This procedure involves employing text semantic similarity, utilizing a similarity calculation method based on optimized Word Mover Distance (WMD), and designing long ICH communication clustering. The main objective of this analysis is to enhance the understanding and classification of the intricate nature of ICH within the musical realm. Finally, experiments are conducted to confirm the model’s effectiveness. The results show that: (1) the text word vector training based on the word2vec model is highly accurate; (2) with the increase in K value, the effect of each category of intangible word vector is improving; (3) the final F1-measure value of the clustering experiment based on the optimized WMD is 0.84. These findings affirm the usefulness and accuracy of the proposed methodology.

Mother-infant social gaze dynamics relate to infant brain activity and word segmentation

The ‘social brain’, consisting of areas sensitive to social information, supposedly gates the mechanisms involved in human language learning. Early preverbal interactions are guided by ostensive signals, such as gaze patterns, which are coordinated across body, brain, and environment. However, little is known about how the infant brain processes social gaze in naturalistic interactions and how this relates to infant language development. During free-play of 9-month-olds with their mothers, we recorded hemodynamic cortical activity of ´social brain` areas (prefrontal cortex, temporo-parietal junctions) via fNIRS, and micro-coded mother’s and infant’s social gaze. Infants’ speech processing was assessed with a word segmentation task. Using joint recurrence quantification analysis, we examined the connection between infants’ ´social brain` activity and the temporal dynamics of social gaze at intrapersonal (i.e., infant’s coordination, maternal coordination) and interpersonal (i.e., dyadic coupling) levels. Regression modeling revealed that intrapersonal dynamics in maternal social gaze (but not infant’s coordination or dyadic coupling) coordinated significantly with infant’s cortical activity. Moreover, recurrence quantification analysis revealed that intrapersonal maternal social gaze dynamics (in terms of entropy) were the best predictor of infants’ word segmentation. The findings support the importance of social interaction in language development, particularly highlighting maternal social gaze dynamics.

Research on the Influence of Socialist Core Value System Construction on College Students’ Ideological and Political Education in the Context of Deep Learning

Abstract In this paper, the word-level N-gram language model applicable to Chinese word segmentation in ideological and political education is obtained through large-scale corpus training, and the Chinese word segmentation path in ideological and political education is calculated according to the conditional probability formula. On the basis of the deep learning algorithm, the Chinese participle algorithm is optimized by introducing word frequency deviation and sorted word frequency deviation indexes, and the cut score with the largest value of the participle indexes is taken as the final participle result output. Starting from the current situation, we determine the evaluation indexes of ideological and political education. Meanwhile, we complete the research design of the impact of the socialist core value system and use the SPSS22.0 and Origin2019 statistical software to carry out the research and analysis of ideological and political education. The results show that the index values of the Chinese word separation algorithm based on the 4-gram language model have also been improved by 1.15%, which means that there exists the possibility of practical application of the model in Chinese word separation in ideological and political education. The college students’ identification with the socialist core values at the national, social and personal levels are all at a high level, with values of 4.28±0.82, 4.32±0.75 and 4.61±0.62 respectively. This study improves students’ ideological-political literacy and provides references to their healthy growth.

Research on Mining Talent Demand for E-commerce Majors based on LDA Topic Model

[Purpose] Based on the text mining method, this study analyzes the job demand for domestic e-commerce industry positions in the Internet-oriented recruitment data, promotes the matching of e-commerce positions and talents, and promotes the good construction of the employment environment in the domestic e-commerce industry.[Method] Use the LDA(Latent dirichlet allocation) topic model for job competency requirements to calculate the similarity of the word segmentation results for professional talent requirements, determine the optimal number of topics, and output the visualization results of the LDA topic model.[Conclusion] In eastern China, there is a high demand for e-commerce jobs, and the majority of these positions only call for 1-3 years of experience. The basic requirement for the majority of jobs is a college degree. The Director of Operations and Director of Network Operations focus on e-commerce operations and management, the Network Promotion Specialist focuses on e-commerce advertising and product promotion, the Data Analyst and Market Analyst focus on data analysis and market research, and the Sales Representative focuses on sales. There is a correlation between the salary of the positions and education, experience and region; the higher the education, the more experience and the higher the level of regional development, the higher the salary level.

Hybrid Arabic handwritten character segmentation using CNN and graph theory algorithm

Arabic handwritten words segmentation is considered a challenging step in Arabic optical character recognition. In this work, we propose a convolutional neural network-based segmentation of handwritten Arabic words. After generating candidate segmentation points, we applied a graph theory-based technique to divide the word into sub-words. Our approach reached a 96% correct sub-word segmentation rate. Additionally, we develop a Segmentation Hypothesis Graph (SHG) to generate candidate characters for each sub-word. We employ both a manually designed convolutional neural network and a transfer learning-based CNN using the pretrained AlexNet model. The accuracy, 96.97%, of the pretrained model outperformed the manually and state of the art models trained on the same dataset. Finally, we used recognition confidence to validate the segmentation points. Hence, our approach correctly segments 88% of the words into characters and successfully resolves 100% of the overlapping cases.

Generalized Segmentation Algorithm for Dissimilar Script Languages

Optical Character Recognition is considered one of the fastest methods of data entry. OCR converts the text image representation of x and y coordinates representing pixel information to be converted into text data in a particular language. OCR as a field of pattern recognition and document image understanding, OCR requires a challenging job once a different language text is available on the image. Difference in language script will pose different challenges for OCR which requires entirely different approaches and algorithms. Latin scripts require a different approach whereas the Arabic adopted language scripts require a different approach. In this regard, various solutions have been proposed for different languages. Segmentation is considered one of the important tasks in the process of OCR. A good segmentation will definitely increase the accuracy of an OCR. Segmentation includes the segmentation of text lines from text images which are further divided into words. These segmented words are further divided into characters which are to be recognized. A single segmentation algorithm to segment various scripts of the languages is proposed in this study which checks the script and then segments the text image for the further processing in OCR. The proposed generalized algorithm will check the style, direction and other properties of the script and then adopts the segmentation process to segment text lines, words and characters of the language. The proposed algorithm segments more than ten languages of three scripts and segments for their OCRs. These images can be further processed for feature extraction and classification further. The process of OCR for selected languages will be made easier to recognize. Multiple scripts, languages and images were experimented, and the proposed algorithm successfully segmented 32,833 images of text line, words and character image. The algorithm provides 97% accuracy while segmenting these images and can be extended to further languages as well as scripts.

Strategies for Monitoring and Managing Online Public Opinion in Universities Under the Background of Big Data

<p>In response to the issue of some keywords not being logged in or having inaccurate semantics in current public opinion monitoring, this article uses an improved word segmentation method to extract semantic features. In response to the current issue of unable to control the emotional direction of public opinion comments in public opinion analysis, an emotion analysis model Bi_GRU is proposed for sentiment analysis. Finally, using students’ commonly used Weibo as a verification scenario, sensitive information such as “food safety” and “campus bullying” is screened to control the emotional direction of college students. The final proof is that the method proposed in this article can effectively supervise public opinion in a centralized environment and provide effective means for student management. </p> <p> </p>

Infants’ early associations between segments of words and referents: Precursors of morphology acquisition

Antecedents and objectiveDuring morphology acquisition, children must learn to segment and associate the morphemes of a word to its meaning or grammatical function, despite the complex morphological variability children may encounter in their linguistic environment. The present study explores the ability of 9–12-month-old infants to segment a morpheme-like syllable of varying pseudowords and associate it to a novel object. Participants and methodThirty-nine Spanish-learning infants participated in an experiment involving a preferential looking task. During each of five training trials per category, infants were presented to one of two novel objects while simultaneously heard two instances of pseudowords with a common morpheme-like segment either at the beginning or the end of the pseudoword (e.g., /tabi/, /babi/, for object 1; or /sato/, /same/, for object 2). Children were then tested in five trials per object. The two trained objects were presented side-by-side while infants heard new pseudowords that followed the pattern of the pseudowords that were heard during training. ResultsResults show an effective association between a referent and a group of pseudowords with a repeating morpheme-like segment. This association is observed both with beginning and final segments. ConclusionsThese findings show infants’ basic abilities for the processing of morphology at an earlier age than it has been suggested. The potential influence of morphologically rich languages such as Spanish is considered.

The Role of Format Familiarity and Word Frequency in Chinese Reading.

For Chinese readers, reading from left to right is the norm, while reading from right to left is unfamiliar. This study comprises two experiments investigating how format familiarity and word frequency affect reading by Chinese people. Experiment 1 examines the roles of format familiarity (reading from left to right is the familiar Chinese format, and reading from right to left is the unfamiliar Chinese format) and word frequency in vocabulary recognition. Forty students read the same Chinese sentences from left to right and from right to left. Target words were divided into high and low frequency words. In Experiment 2, participants engaged in right-to-left reading training for 10 days to test whether their right-to-left reading performance could be improved. The study yields several main findings. First, format familiarity affects vocabulary recognition. Participants reading from left to right had shorter fixation times, higher skipping rates, and viewing positions closer to word center.. Second, word frequency affects vocabulary recognition in Chinese reading. Third, right-to-left reading training could improve reading performance. In the early indexes, the interaction effect of format familiarity and word frequency was significant. There was also a significant word-frequency effect from left to right but not from right to left. Therefore, word segmentation and vocabulary recognition may be sequential in Chinese reading.

Fine-Grained Domain Adaptation for Chinese Syntactic Processing

Syntactic processing is fundamental to natural language processing. It provides rich and comprehensive syntax information in sentences that could be potentially beneficial for downstream tasks. Recently, pretrained language models have shown great success in Chinese syntactic processing, which typically involves word segmentation, POS tagging, and dependency parsing. However, the on-going research never ends since performance would be degraded drastically when tested on a highly-discrepant domain. This problem is widely accepted as domain adaptation, where the test domain differs from the training domain in supervised learning. Self-training is one promising solution for it, and straightforward source-to-target adaptation has already shown remarkable effectiveness in previous work. While this strategy ignores the fact that sentences of the target domain sentences may have very different gaps from the source training domain. More specifically, sentences with large gaps might fail by direct self-training adaptation. To this end, we propose fine-grained domain adaptation for Chinese syntactic processing in this work, aiming to model the gaps between the source and the target domains accurately and progressively. The key idea is to divide the target domain into fine-grained subdomains by using a specified domain distance metric, and then perform gradual self-training on the subdomains. We further offer an intuitive theoretical illustration based on the theory of Kumar et al. (2020) approximately. In addition, a novel representation learning framework is proposed to encode fine-grained subdomains effectively, aiming to utilize the above idea fully. Experimental results on benchmark datasets show that our method can achieve significant improvements over a variety of baselines.

Brain bases of morphological awareness and longitudinal word reading outcomes

Children’s spoken language skills are essential to the development of the “reading brain,” or the neurocognitive systems that underlie successful literacy. Morphological awareness, or sensitivity to the smallest units of meaning, is a language skill that facilitates fluent recognition of meaning in print. Yet despite the growing evidence that morphology is integral to literacy success, associations among morphological awareness, literacy acquisition, and brain development remain largely unexplored. To address this gap, we conducted a longitudinal investigation with 75 elementary school children (5–11 years of age) who completed an auditory morphological awareness neuroimaging task at Time 1 as well as literacy assessments at both Time 1 and Time 2 (1.5 years later). Findings reveal longitudinal brain–behavior associations between morphological processing at Time 1 and reading outcomes at Time 2. First, activation in superior temporal brain regions involved in word segmentation was associated with both future reading skill and steeper reading gains over time. Second, a wider array of brain regions across the language network were associated with polymorphemic word reading as compared with broader word reading skill (reading both simple and complex words). Together, these findings reinforce the importance of word segmentation skills in learning to read and highlight the importance of considering complex word reading skills in building comprehensive neurocognitive models of literacy. This study fills a gap in our knowledge of how processing meaningful units in speech may help to explain differences in children’s reading development over time and informs ongoing theoretical questions about the role of morphology in learning to read.

Statistical word segmentation succeeds given the minimal amount of exposure.

One of the first tasks in language acquisition is word segmentation, a process to extract word forms from continuous speech streams. Statistical approaches to word segmentation have been shown to be a powerful mechanism, in which word boundaries are inferred from sequence statistics. This approach requires the learner to represent the frequency of units from syllable sequences, though accounts differ on how much statistical exposure is required. In this study, we examined the computational limit with which words can be extracted from continuous sequences. First, we discussed why two occurrences of a word in a continuous sequence is the computational lower limit for this word to be statistically defined. Next, we created shortsyllable sequences that contained certain words either two or four times. Learners were presented with these syllable sequences one at a time, immediately followed by a test of the novel words from these sequences. We found that, with the computationally minimal amount of two exposures, words were successfully segmented from continuous sequences. Moreover, longer syllable sequences providing four exposures to words generated more robust learning results. The implications of these results are discussed in terms of how learners segment and store the word candidates from continuous sequences.

Segmentation of Spoken Overlapping Ambiguity Strings in Chinese: An Eye-Tracking Study.

Using the printed-word paradigm with eye tracking, this study conducted three experiments to examine (a) how multiple words in spoken overlapping ambiguity strings (OASs) are activated, (b) how word frequency influences the word segmentation of spoken OASs, and (c) whether the multiple words in spoken OASs are activated competitively or independently. In this study, participants listened to a four-character spoken OAS (ABCD) and were presented with a visual display composed of a semantic associate of the "middle word" (BC; Experiments 1 and 2) or the "left word" (AB; Experiment 3) and two distractors. In Experiment 1, the word frequency of the middle words was manipulated to be higher than that of the neighbor words. In Experiment 2, the word frequency of the middle words was manipulated to be either higher or lower than that of the neighbor words. In Experiment 3, participants listened to either spoken OASs (ABCD) or spoken unambiguous strings (ABEF). In Experiment 1, we observed a significant semantic competition effect; that is, more fixations fell on the semantic competitors than on distractors, suggesting that the semantic information of the middle words in the spoken OASs was activated. In Experiment 2, the semantic competition effect was only observed in the high-frequency condition and was absent in the low-frequency condition. In Experiment 3, the results showed significant semantic competition effects for the left words under both conditions, and the observed effect was similar between the ambiguity condition and the control condition. These findings show that multiple words in spoken OASs are all activated and the activation level is modulated by word frequency. In addition, multiple words in the spoken OASs may be processed independently during spoken comprehension.

Machine learning and data analysis for word segmentation of classical Chinese poems: illustrations with Tang and Song examples

Abstract Words are essential parts for understanding classical Chinese poems. We report a collection of 32,399 classical Chinese poems that were annotated with word boundaries. Statistics about the annotated poems support a few heuristic experiences, including the patterns of lines and a practice for the parallel structures (對仗), that researchers of Chinese literature discuss in the literature. The annotators were affiliated with two universities, so they could annotate the poems as independently as possible. Results of an inter-rater agreement study indicate that the annotators have consensus over the identified words 93 per cent of the time and have perfect consensus for the segmentation of a poem 42 per cent of the time. We applied unsupervised classification methods to annotate the poems in several different settings, and evaluated the results with human annotations. Under favorable conditions, the classifier identified about 88 per cent of the words, and segmented poems perfectly 22 per cent of the time.

A multi-dimensional fusion strategy similarity measure method for patent application technology disclosure document.

Patent application technology disclosure document is one of the important bases for judging patent novelty and uniqueness. Automated evaluation can effectively solve the problems of long time and strong subjectivity of human evaluation. The text similarity evaluation algorithm based on corpus and deep learning technology has problems such as insufficient amount of cross-library learning data and insufficient core content tendency in the similarity judgment of patent application technology disclosure document, which limits their performance and practical application. In this paper, we propose a similarity evaluation method of patent application technology disclosure document based on multi-dimensional fusion strategy to realize the similarity measurement of patents. Firstly, in the text preprocessing section, word segmentation reconstruction and similarity evaluation optimization strategies based on word frequency and part-of-speech score weighted fusion are proposed. Then, a similarity calculation method of patent application technology disclosure document based on two new mapping spaces of dot matrix and image is proposed to achieve a more diversified comprehensive evaluation. The algorithm was evaluated by using four published text similarity matching datasets (containing 0-5 or 0/1 labels) and a set of patent application technology disclosure documents. Experimental results show that on the published text similarity matching datasets, the similarity evaluation method under the multi-dimensional fusion strategy proposed in this paper has a discrimination accuracy improvement of about 10% compared to traditional vector semantic model, and can match the discriminative ability of lightweight deep learning models without the need for training. At the same time, the discrimination accuracy of the proposed method on the sample dataset of patent application technology disclosure document is superior to traditional vector semantic model (20%) and various deep learning models (1%-8%), and the precision and recall rate are relatively balanced. The visual analysis results on the dataset of the patent application technology disclosure documents also prove the effectiveness and reliability of the similarity calculation method proposed in the dot matrix and image space, which provide a new idea and method for the similarity evaluation between patent application technology disclosure document.