Chinese Classification Research Articles

Prevalence of obesity and associated complications in China: A cross-sectional, real-world study in 15.8 million adults.

To evaluate the prevalence of overweight/obesity and associated complications from a large, cross-sectional, nationwide database in China. Data were obtained from 519 Meinian health check-up centres across 243 cities. Eligible participants were aged ≥18 years, with a routine check-up in 2019 (N = 21 771 683) and complete height, weight, sex and region data. The unadjusted prevalence rates of overweight/obesity were calculated by age, sex and region. In addition, the nationwide prevalence rates of overweight and obesity were standardized according to the 2010 China census by age group and sex. The prevalence of obesity-related complications by body mass index (BMI) groups was calculated using logistic regression. There were 15 770 094 eligible participants (median age 40 years; mean BMI 24.1 kg/m2 ; 52.8% male). By Chinese BMI classification, 34.8% were overweight and 14.1% were obese. Overweight and obesity were more prevalent in male than female participants (standardized: overweight 40.2% vs. 27.4%; obesity 17.6% vs. 9.6%, respectively). The prevalence of assessed complications was higher in participants with overweight/obesity versus those with normal BMI (P < 0.001 for trends). The most prevalent complications in participants with overweight/obesity were fatty liver disease, prediabetes, dyslipidaemia and hypertension. The number of complications increased with higher BMI. Overweight/obesity and related complications are highly prevalent in this population. These data may better inform management and prevention public health strategies in China.

An adversarial-example generation method for Chinese sentiment tendency classification based on audiovisual confusion and contextual association

An adversarial-example generation method for Chinese sentiment tendency classification based on audiovisual confusion and contextual association

VS30-based relationship for Chinese site classification

Classification of engineering sites is essential when considering site conditions for seismic fortification in seismic design codes. However, because of different site classification methods, it is difficult for Chinese seismological and engineering communities to share site metadata (e.g., site classes) and relevant results with international communities. This issue also prevents the Chinese communities from utilizing well-established models based on the time-averaged shear wave velocity for the top 30 m (VS30). This study aims to close the gap and develop a consistent VS30-based site classification method. We first assemble the shear wave velocity (VS) profile data from various sites in China, Japan, and California to study the statistical characteristics of VS30 and the equivalent shear wave velocity (VSe, time-averaged within soil layers for ≤ 20 m, which is used in Chinese site classification). We then investigate the efficiency of site classification by VS30 and VSe alone, respectively. We find that the single VSe parameter is insufficient to distinguish different Chinese site classes, especially for site classes III and IV (which consider soil layers deeper than 20 m). In contrast, the VS30 parameter can effectively differentiate Chinese site classes from I0 to IV. A new VS30-based site classification method for Chinese sites is then developed. The boundaries of VS30 for each Chinese site class are determined by maximizing the rate of correctness. Our new VS30-based classification facilitates the connection of Chinese site classes with well-developed VS30-based models and site data sharing between Chinese and international scientific communities.

The value of serum Lp-PLA2 combined with MPO in the diagnosis of cerebral infarction caused by large artery atherosclerosis

ObjectiveTo explore the value of serum lipoprotein-associated phospholipase A2（Lp-PLA2）combined with myeloperoxidase（MPO）for the diagnosis of large artery atherosclerosis(LAA) cerebral infarction. MethodsBaseline data were collected from patients with first-ever acute cerebral infarction, serum Lp-PLA2 and MPO levels were measured. The etiology of cerebral infarction was classified according to the Chinese Ischemic Stroke Subtype Classification Standard. The risk factors associated with LAA cerebral infarction were identified by univariate and multivariate regression analysis. The diagnostic value of serum Lp-PLA2 and MPO for LAA cerebral infarction was assessed by the area under the receiver-operating characteristic (ROC) curve. ResultsOverall 368 patients were involved, 148 patients (40.22 %) were LAA. The serum La-PLA2 and MPO levels were higher in the LAA group than those in non-LAA group (23.06 ± 3.39 ng/mL versus 17.48 ± 3.26 ng/mL; 93.60 ± 9.58 ng/mL versus 75.98 ± 15.53 ng/mL; P < 0.001 for both). Multivariate analysis showed that elevated levels of serum Lp-PLA2 (OR 1.742, 95 %CI 1.499–2.025; P < 0.001) and MPO (OR 1.060, 95 % CI 1.026–1.096; P = 0.001) were the independent risk factors of LAA cerebral infarction. The area under curve of the serum Lp-PLA2 combined with MPO for the diagnosis of LAA cerebral infarction was 0.896 [0.866 ∼ 0.927] (P < 0.001). ConclusionSerum Lp-PLA2 combined with MPO could be valued as a predictor of acute cerebral infarction caused by large artery atherosclerosis.

Ontology alignment with semantic and structural embeddings

Ontology alignment is essential for data integration and interoperability across multiple applications across diverse disciplines. In recent decades, significant advancements have been made in the development of advanced methods and systems for ontology alignment. Empirical results have suggested that ontological semantics can be effectively employed to enhance the alignment process. Besides, structural information is crucial for ontology alignment as it reflects the relations among adjacent concepts in the ontology. Previous works are mainly based on external lexicon and predefined rules based on ontological structure. Recently, deep learning has imposed positive impacts on ontology alignment and obtained substantial improvement.This paper proposes a new method based on ontology embedding incorporating the semantic and structural features. It utilizes the distance between the embedding of two ontological concepts to be aligned as the criterion for alignment. The proposed method is used to align two widely used food ontologies and three Chinese food classification ontologies. The experimental results show that our method enhances the performance compared to several state-of-the-art alignment systems, demonstrating the importance of learning semantic representation and structural representation. Furthermore, the proposed method is evaluated on several different tracks of the Ontology Alignment Evaluation Initiative (OAEI), and experimental results show that our method outperforms other baselines in effectiveness. The data and code can be obtained from: https://github.com/haozhigang1111/Ontology-Alignment.git.

A Fine-grained Chinese Short Text Classification Method Based on Capsule Networks

Chinese short-text classification is an essential primary research direction in natural language processing. The main problem facing the current Chinese short text classification task is how to effectively extract critical semantic feature information and determine the text category within a short text length. Aiming at the problem of feature information extraction of Chinese short texts, this paper introduces a novel sparse attention mechanism. It proposes a Chinese short text classification model RGSC combined with capsule networks. After the model generates word vectors through the RoBERTa pre-training model, the Bi-GRU model initially extracts text feature information. The profound text feature extraction module SC, composed of sparse attention mechanism and capsule networks, is used to further extract critical semantic feature information. The results of ablation experiments on the TouTiao Chinese short text dataset and performance comparison experiments with various models show that the RGSC model can effectively extract the essential text feature information while effectively reducing the irrelevant noise information contained in the text features, and obtain fine-grained text feature information.

On Chinese Children’s Acquisition of Motion Events

The motion events expressions in Chinese have been a controversial topic and few research on them from the developmental perspective. Therefore, based on the long-term tracking corpus of two Mandarin children SWK and WMX from their 13 months to 48 months, this paper studies the lexicalization of Chinese children’s motion events in the process of language acquisition to reveal the unique features in Chinese motion event expressions. Research revealed that at age 3, Chinese children began to express motion events primarily using the serial verb structure, that is, a manner verb + path verb pattern and children’s motion event expressions grew progressively more extensive with age. Chinese children follow the satellite-framed language pattern early on. The authentic language use in corpora and the features found based on them prove that Chinese is a satellite-framed language. The present study offers a new type of evidence for the typological properties of the language from the developmental perspective and could shed fresh insight on a thorny issue regarding the classification of Mandarin Chinese in motion events.

Deep Chnet 1-D CNN using raw audio waveform for recognizing traditional Chinese musical instrument

The classification of music information using various deep learning models is increasingly popular in the field of Music Information Retrieval research. However, as most proposed works focus on western music and musical instruments, little attention is given to traditional Chinese music. This paper proposes a 1-D Convolutional Neural Network (1-D CNN) using only raw audio waveform as input, to undertake the task of traditional Chinese musical instruments classification. This paper starts with a review of the current state of research on the related field, then discuss the proposed model and its data in detail, followed by its performance metrics and then a conclusion on the experiment. The result shows that 1-D CNN provides competitive and even superior results when compared to its 2-D versions as well as when compared to traditional models.

Feature fusion and decomposition: exploring a new way for Chinese calligraphy style classification

Feature fusion and decomposition: exploring a new way for Chinese calligraphy style classification

A Study of Chinese News Headline Classification Based on Keyword Feature Expansion

Existing work generally classifies news headlines as a matter of short text classification. However, due to the strong domain nature and limited text length of news headlines, their classification results are usually determined by several specific keywords, which makes the traditional short text classification method ineffective. In this paper, we propose a new method to identify keywords in news headlines and expand their features from sentence level and word level respectively, and finally use convolutional neural networks (CNN) to extract and classify their features. The proposed model was tested on the Sogou News Corpus dataset and achieved 93.42% accuracy.

Associations of maternal pre-pregnancy body mass index and gestational weight gain with risk of offspring neurodevelopment at 2years: A Chinese birth cohortstudy.

In recent decades, there has been a surge in both obesity and developmental impairments. Only a few research have looked at the relationship between gestational weight growth and pre-pregnancy BMI in mothers and the neurobehavioral development of their infants. The current research investigates the associations among maternal pre-pregnancy BMI, GWG, and the risk of child neural development at 2 years of age depending on a Chinese birth prospective study. The study population was 3,115 mother-infant pairs were registered in the Wuhan Health Baby cohort between September 2013 and October 2018, and data from this cohort was used in this investigation. The Chinese classification was used to group maternal BMI before conception. Based on the 2019 Life Cycle Project-Maternal Obesity and Childhood Outcomes Study Group, categories for GWG were created. The outcome was an assessment of child neural development at age 2 which was measured by employing a Chinese translation of the Bayley scales (BSID-CR). The multivariate regression models were used to calculate the beta (β) coefficients and 95% confidence intervals (CIs) for estimating the associations between continuous Bayley scores and maternal pre-pregnancy BMI categories, as same as in GWG categories. Infants of overweight and obese moms exhibited lower MDI scores than those of mothers with normal pre-pregnancy BMI (β = -2.510, 95%CI = -4.821 to -0.200) in the entire sample. Meanwhile, we find among the normal pre-pregnancy BMI mothers, infants of inadequate GWG mothers had lower MDI scores (β = -3.952, 95%CI = -7.809 to -0.094) compared with the referenced adequate GWG mothers, as well as the infants of excessive GWG mothers among the underweight pre-pregnancy BMI mothers (β = -5.173, 95%CI = -9.803 to -0.543). The PDI scores of the infants were not affected by the maternal pre-pregnancy BMI or GWG. For Chinese babies aged 2 in this nationally representative sample, aberrant pre-pregnancy BMI and GWG can impair infants' mental development, but not psychomotor development. Such results are significant given the incidence of overweight and obesity as well as the long-term effects of early brain development. In this study we found optimal GWG recommendations proposed by 2019 Life Cycle Project-Maternal Obesity and Childhood Outcomes Study Group were more suitable for Chinese women than 2009 Institute of Medicine(IOM) guidelines. Additionally, women should be given general advice on how to achieve their ideal pre-pregnancy BMI and GWG.

Interpretation of Chinese expert consensus on the pelvic exenteration for primary rectal cancer beyond total mesorectal excision planes and locally recurrent rectal cancer (2023 edition)

The China PelvEx Collaborative, under the direction of Colorectal Cancer Committee of the Chinese Medical Doctor Association, Gastrointestinal Surgery Committee of China International Exchange and Promotive Association for Medical and Health, has formulated and issued the Chinese expert consensus for primary rectal cancer beyond total mesorectal excision planes and locally recurrent rectal cancer (2023 edition) , with the academic support of the Chinese Journal of Gastrointestinal Surgery and Chinese Journal of Colorectal Disease (Electronic Edition). This Consensus refers to the expert consensus developed by the International PelvEx Collaboration, incorporates the latest international multi-center research results and combines the latest research results in China. The Consensus unifies some definitions, clarifies the surgical indications, and puts forward the definition and preventive measures of "empty pelvic syndrome" earlier. For the controversial classification of local recurrent rectal cancer, the Chinese classification was proposed for the first time in Consensus. At the same time, the definition of pelvic exenteration is controversial, and a more consistent cognition is proposed. It is believed that, with the in-depth research on complicated rectal cancer, C-PelvEx will gather more higher-level data from clinical research in several domestic centers, so as to further enrich the content of the updated Consensus.

Research on Chinese Emotion Classification using BERT-RCNN-ATT

Emotional classification is the process of analyzing and reasoning subjective texts with emotional color, that is, analyzing whether their emotional tendencies are positive or negative. Aiming at the problems of massive data and nonstandard words in the existing Chinese short text emotion classification algorithm, the traditional BERT model does not distinguish the semantics of words with the same sentence pattern clearly, the multi-level transformer training is slow, time-consuming, and requires high energy consumption, this paper proposes to classify users' emotions based on BERT-RCNN-ATT model, and extract text features in depth using RCNN combined with attention mechanism, Multi task learning is used to improve the accuracy and generalization ability of model classification. The experimental results show that the proposed model can more accurately understand and convey semantic information than the traditional model. The test results show that compared with the traditional CNN, LSTM, GRU models, the accuracy of text emotion recognition is improved by at least 4.558%, the recall rate is increased by more than 5.69%, and the F1 value is increased by more than 5.324%, which is conducive to the sustainable development of emotion intelligence combining Chinese emotion classification with AI technology.

A model of integrating convolution and BiGRU dual-channel mechanism for Chinese medical text classifications.

Recently, a lot of Chinese patients consult treatment plans through social networking platforms, but the Chinese medical text contains rich information, including a large number of medical nomenclatures and symptom descriptions. How to build an intelligence model to automatically classify the text information consulted by patients and recommend the correct department for patients is very important. In order to address the problem of insufficient feature extraction from Chinese medical text and low accuracy, this paper proposes a dual channel Chinese medical text classification model. The model extracts feature of Chinese medical text at different granularity, comprehensively and accurately obtains effective feature information, and finally recommends departments for patients according to text classification. One channel of the model focuses on medical nomenclatures, symptoms and other words related to hospital departments, gives different weights, calculates corresponding feature vectors with convolution kernels of different sizes, and then obtains local text representation. The other channel uses the BiGRU network and attention mechanism to obtain text representation, highlighting the important information of the whole sentence, that is, global text representation. Finally, the model uses full connection layer to combine the representation vectors of the two channels, and uses Softmax classifier for classification. The experimental results show that the accuracy, recall and F1-score of the model are improved by 10.65%, 8.94% and 11.62% respectively compared with the baseline models in average, which proves that our model has better performance and robustness.

Chinese Spam Classification based on Deep Learning

In the 21st century, it is an information age where individuals and organizations use computers to work and they commonly use email to chat with each other, share knowledge or their wishes or discuss business. However, in this context, spam is harassing our lives. Spam is a type of email that is delivered to lots of numbers of recipients without asking them for their wishes. Spam wastes our time reading useless messages and takes up our email addresses and network traffic. The problem of spam has become increasingly important and is growing over the years. The emergence of machine learning technology can simulate human intelligence. Use machine learning to identify the construct of pictures and words. Through this technology, this paper can classify the kind of different messages. Spam is very common nowadays; it can easily send lots of message to occupying a large amount of network bandwidth. Access such kind of messages, this article uses four models: Decision Tree, Support Vector Machine (SVM), Naive Bayes model, and Logistic Regression to identify spam to prevent it into our electronic mail.

A Comparative Study of Convolutional Neural Network Model for Chinese Flower Tea Classification

While flower teas are well-known for their health benefit, little did people know, there are several types of flower tea, and each type has its health benefit. Due to the unavailability of an automated system for classifying Chinese flower tea at the meantime, we then decided to apply the Convolutional Neural Network to help the wider community or flower tea plantation owners to classify flower tea more quickly, accurately, and automated. The purpose of this research is to classify flower tea based on their type by using CNN algorithm. In this research, we used multiple CNN models to find the most suitable architecture. The CNN models compared are ResNet50, SqueezeNet, AlexNet, and ResNet18. The result indicates AlexNet to achieve the highest accuracy of 97.92%

Image classification of Chinese medicinal flowers based on convolutional neural network.

Traditional Chinese medicine has used many herbs on the prevention and treatment of diseases for thousands of years. However, many flowers are poisonous and only few herbs have medicinal properties. Relying on experts for herbs identification is time consuming. An efficient and fast identification method is proposed in this study. This study proposes ResNet101 models by combining SENet and ResNet101, adding convolutional block attention module or using Bayesian optimization on Chinese medicinal flower classification. The performances of the proposed ResNet101 models were compared. The best performance for accuracy, precision, recall, F1-score and PR-AUC are coming from ResNet101 model with Bayesian optimization which are 97.64%, 97.99%, 97.86%, 97.82% and 99.72%, respectively. The proposed ResNet101 model provides a better solution on the image classification of Chinese medical flowers with favourable accuracy.

『字類註釋』의 同字異釋형 중출 표제한자의 새김과 분류 - 한자 새김 자료와 비교를 중심으로 -

본고에서는 19세기의 문헌 『자류주석』의 특징적 새김 양상인 동자이석(同字異釋)형 중출 표제한자의 새김과 수록 방식을 살펴보았다. 한자 새김 자료에서 표제한자의 중복 출현은 매우 드물다. 본고에서 분석한 『자류주석』의 동자이석 중출 표제한자는 48개이다. 대다수가 한자초학서에는 나오지 않는데, 20세기의 자서류 『자전석요』, 『신자전』의 풀이와 비교해 보면, 『자전석요』와 『신자전』에는 모두 단일 표제한자로 수록되어 복수의 새김이 제시되어 있다. 『자류주석』과 『신자전』의 새김 방식이 유사하고, 한문주석이 공통적인 부분이 많다. 『자류주석』은 이전 한자초학서의 정보를 단순 답습하지 않고 여러 전거 문헌을 면밀히 살펴 개별 한자의 새김을 정리하고 이를 의미 부류 분류를 정밀히 하여 표제한자로 중복 제시하였다. 이로써 분류어휘집의 정체성을 더욱 분명히 하는 동시에 개별 한자에 대한 이해를 심층화하였다.

E3W—A Combined Model Based on GreedySoup Weighting Strategy for Chinese Agricultural News Classification

With the continuous development of the internet and big data, modernization and informatization are rapidly being realized in the agricultural field. In this line, the volume of agricultural news is also increasing. This explosion of agricultural news has made accurate access to agricultural news difficult, and the spread of news about some agricultural technologies has slowed down, resulting in certain hindrance to the development of agriculture. To address this problem, we apply NLP to agricultural news texts to classify the agricultural news, in order to ultimately improve the efficiency of agricultural news dissemination. We propose a classification model based on ERNIE + DPCNN, ERNIE, EGC, and Word2Vec + TextCNN as sub-models for Chinese short-agriculture text classification (E3W), utilizing the GreedySoup weighting strategy and multi-model combination; specifically, E3W consists of four sub-models, the output of which is processed using the GreedySoup weighting strategy. In the E3W model, we divide the classification process into two steps: in the first step, the text is passed through the four independent sub-models to obtain an initial classification result given by each sub-model; in the second step, the model considers the relationship between the initial classification result and the sub-models, and assigns weights to this initial classification result. The final category with the highest weight is used as the output of E3W. To fully evaluate the effectiveness of the E3W model, the accuracy, precision, recall, and F1-score are used as evaluation metrics in this paper. We conduct multiple sets of comparative experiments on a self-constructed agricultural data set, comparing E3W and its sub-models, as well as performing ablation experiments. The results demonstrate that the E3W model can improve the average accuracy by 1.02%, the average precision by 1.62%, the average recall by 1.21%, and the average F1-score by 1.02%. Overall, E3W can achieve state-of-the-art performance in Chinese agricultural news classification.

Chinese Text Sentiment Classification Based on ERNIE and BiLSTM-AT

Chinese text sentiment classification is a sub-task of natural language processing. However, when text representation is carried out, the polysemy of a word cannot be processed when using the traditional language model to construct the word vector, and the long-distance text information cannot be fully extracted when extracting text features. To solve this problem, this paper proposes a text sentiment classification model combining ERNIE and BiLSTM-AT. First, the pre-training model ERNIE is used to obtain the word vector representation of the fused statement context. Then, the bidirectional long-short-term memory neural network is used to extract the context information and depth semantic information of the text. Then, the attention mechanism is used to assign the corresponding weights to the hidden layer vectors of each time step output by the BiLSTM layer, and the weighted summation is integrated into the sentence features. Finally, the softmax function is used to calculate the probability distribution of the emotional category of the text in the output layer. The results show that the proposed model can achieve high accuracy on both hotel reviews and takeaway reviews. Based on the pre-training model, adding bidirectional long-term and short-term memory network and attention mechanism is beneficial to improve the classification effect of the model, and has certain practicability in text sentiment classification tasks.