RESUMO
Text pre-processing is an important component of a Chinese text classification. At present, however, most of the studies on this topic focus on exploring the influence of preprocessing methods on a few text classification algorithms using English text. In this paper we experimentally compared fifteen commonly used classifiers on two Chinese datasets using three widely used Chinese preprocessing methods that include word segmentation, Chinese specific stop word removal, and Chinese specific symbol removal. We then explored the influence of the preprocessing methods on the final classifications according to various conditions such as classification evaluation, combination style, and classifier selection. Finally, we conducted a battery of various additional experiments, and found that most of the classifiers improved in performance after proper preprocessing was applied. Our general conclusion is that the systematic use of preprocessing methods can have a positive impact on the classification of Chinese short text, using classification evaluation such as macro-F1, combination of preprocessing methods such as word segmentation, Chinese specific stop word and symbol removal, and classifier selection such as machine and deep learning models. We find that the best macro-f1s for categorizing text for the two datasets are 92.13% and 91.99%, which represent improvements of 0.3% and 2%, respectively over the compared baselines.
Assuntos
População do Leste Asiático , Envio de Mensagens de Texto , Humanos , AlgoritmosRESUMO
BACKGROUND: Edible flowers (EFs) represent valuable sources of both food and medicinal resources, holding the promise to enhance human well-being. Unfortunately, their significance is often overlooked. Ethnobotanical studies on the EFs are lacking in comparison with their botanical and phytochemical research. The practice of consuming flowers as food has a rich culture and long history in China, especially among different linguistic groups in Xishuangbanna, Yunnan. However, economic activities have led to a decline of this tradition. Consequently, preserving the traditional knowledge and culture tied to the EFs in Xishuangbanna becomes both essential and pressing. METHODS: The field ethnobotanical survey was conducted in Xishuangbanna during five visits in April 2021 and May 2023, covering 48 villages and 19 local markets of all three county-level areas and 9 different linguistic groups. By conducting a comprehensive literature review and on-site field surveys, relevant information regarding the EFs of Xishuangbanna was systematically collected and documented. Additionally, the relative frequency of citation (RFC) values were calculated from the survey data. RESULTS: A total of 212 taxa (including species and varieties) of EFs from 58 families and 141 genera were documented in the study area. The edible parts of flowers were classified into 13 categories including peduncle, petal, flower buds, inflorescence as a whole, and etc. They were consumed in 21 ways and as 8 types of food. The inflorescence was the most commonly consumed category, accounting for 85 species (40.1%) of the total categories. They always eat flowers as vegetables (184 species, 86.8%). The preparing form of stir-frying was the preferred food preparation method (138, 65.1%). The Xishuangbanna locals had profound knowledge of which EFs required specific processing to remove their toxicity or bitterness. The dishes can be made from either exclusively from the flowers themselves or by incorporating them alongside other plant parts like stems and leaves. Some EFs with high RFC value, such as Musa acuminata and Bauhinia variegata var. candida, showed significant cultural meanings. These edible flowers occupy specific positions in local traditional culture. CONCLUSION: Traditional knowledge regarding edible flowers holds substantial significance and serves as a representative element of the flower-eating culture in Xishuangbanna. Nevertheless, this knowledge and cultural practice are currently decreasing. Serving as a bridge between tradition and modernity, the flower-eating culture, which derives from local people's practical experience, shows the potential of EFs and can be applied to the conservation of biocultural diversity, healthy food systems, and sustainable development.
Assuntos
Etnobotânica , Verduras , Humanos , China , Etnobotânica/métodos , Inquéritos e Questionários , Flores , Plantas ComestíveisRESUMO
The ongoing emergence of COVID-19 and the maturation of cold chain technology, have aided in the rapid development of the fresh produce e-commerce industry. Taking into account the characteristics of consumers' demand for fresh products, this paper constructs a location allocation model of a front warehouse for fresh e-commerce with the objective of minimizing the total cost. An improved immune optimization algorithm is proposed in this paper, and the effectiveness of the proposed algorithm is demonstrated by a real case study. The results show that the improved immune optimization algorithm outperforms the traditional genetic algorithm in terms of solution accuracy; the proposed location model can effectively help fresh produce e-commerce enterprises open new front-end warehouses when demand is increasing, as well as provide optimal economic decision-making for front warehouse layout.
Assuntos
COVID-19 , Humanos , COVID-19/epidemiologia , Comércio , Indústrias , AlgoritmosRESUMO
Recognition of Traditional Chinese Medicine (TCM) entities from different types of literature is challenging research, which is the foundation for extracting a large amount of TCM knowledge existing in unstructured texts into structured formats. The lack of large-scale annotated data makes unsatisfactory application of conventional deep learning models in TCM text knowledge extraction. Some other unsupervised methods rely on other auxiliary data, such as domain dictionaries. We propose a multigranularity text-driven NER model based on Conditional Generation Adversarial Network (MT-CGAN) to implement TCM NER with small-scale annotated corpus. In the model, a multigranularity text features encoder (MTFE) is designed to extract rich semantic and grammatical information from multiple dimensions of TCM texts. By differentiating the conditional constraints of the generator and discriminator of MT-CGAN, the synchronization between the generated tag labs and the named entities is guaranteed. Furthermore, seeds of different TCM text types are introduced into our model to improve the precision of NER. We compare our method with other baseline methods to illustrate the effectiveness of our method on 4 kinds of gold-standard datasets. The experiment results show that the standard precision, recall, and F1 score of our method are higher than the state-of-the-art methods by 0.24â¼8.97%, 0.89â¼12.74%, and 0.01â¼10.84%. MT-CGAN is able to extract entities from different types of TCM literature effectively. Our experimental results indicate that the proposed approach has a clear advantage in processing TCM texts with more entity types, higher sparsity, less regular features, and a small-scale corpus.
Assuntos
Medicina Tradicional Chinesa , SemânticaRESUMO
Obtaining massive amounts of training data is often crucial for computer-assisted diagnosis using deep learning. Unfortunately, patient data is often small due to varied constraints. We develop a new approach to extract significant features from a small clinical gait analysis dataset to improve computer-assisted diagnosis of Chronic Ankle Instability (CAI) patients. In this paper, we present an approach for augmenting spatiotemporal and kinematic characteristics using the Dual Generative Adversarial Networks (Dual-GAN) to train a series of modified Long Short-Term Memory (LSTM) detection models making the training process more data-efficient. Namely, we use LSTM-, LSTM-Fully Convolutional Networks (FCN)-, and Convolutional LSTM-based detection models to identify the patients with CAI. The Dual-GAN enables the synthesized data to approximate the real data distribution visualized by the t-distributed Stochastic Neighbor Embedding (t-SNE) algorithm. Then we trained the proposed detection models using real data collected from a controlled laboratory study and mixed data from real and synthesized gait features. The detection models were tested in real data to validate the positive role in data augmentation as well as to demonstrate the capability and effectiveness of the modified LSTM algorithm for CAI detection using spatiotemporal and kinematic characteristics in walking. Dual-GAN generated efficient spatiotemporal and kinematic characteristics to augment the training set promoting the performance of CAI detection and the modified LSTM algorithm yielded an enhanced classification outcome to identify those CAI patients from a group of control subjects based on gait analysis data than any previous reports.
Assuntos
Tornozelo , Marcha , Algoritmos , Fenômenos Biomecânicos , Humanos , CaminhadaRESUMO
At present, short text classification is a hot topic in the area of natural language processing. Due to the sparseness and irregularity of short text, the task of short text classification still faces great challenges. In this paper, we propose a new classification model from the aspects of short text representation, global feature extraction and local feature extraction. We use convolutional networks to extract shallow features from short text vectorization, and introduce a multi-level semantic extraction framework. It uses BiLSTM as the encoding layer while the attention mechanism and normalization are used as the interaction layer. Finally, we concatenate the convolution feature vector and semantic results of the semantic framework. After several rounds of feature integration, the framework improves the quality of the feature representation. Combined with the capsule network, we obtain high-level local information by dynamic routing and then squash them. In addition, we explore the optimal depth of semantic feature extraction for short text based on a multi-level semantic framework. We utilized four benchmark datasets to demonstrate that our model provides comparable results. The experimental results show that the accuracy of SUBJ, TREC, MR and ProcCons are 93.8%, 91.94%, 82.81% and 98.43%, respectively, which verifies that our model has greatly improves classification accuracy and model robustness.
RESUMO
Multitask learning (MTL) is an open and challenging problem in various real-world applications, such as recommendation systems, natural language processing, and computer vision. The typical way of conducting multitask learning is establishing some global parameter sharing mechanism among all tasks or assigning each task an individual set of parameters with cross-connections between tasks. However, for most existing approaches, the raw features are abstracted step by step, semantic information is mined from input space, and matching relation features are not introduced into the model. To solve the above problems, we propose a novel MMOE-match network to model the matches between medical cases and syndrome elements and introduce the recommendation algorithm into traditional Chinese medicine (TCM) study. Accurate medical record recommendation is significant for intelligent medical treatment. Ranking algorithms can be introduced in multi-TCM scenarios, such as syndrome element recommendation, symptom recommendation, and drug prescription recommendation. The recommendation system includes two main stages: recalling and ranking. The core of recalling and ranking is a two-tower matching network and multitask learning. MMOE-match combines the advantages of recalling and ranking model to design a new network. Furtherly, we try to take the matching network output as the input of multitask learning and compare the matching features designed by the manual. The data show that our model can bring significant positive benefits.
Assuntos
Medicina Tradicional Chinesa , Processamento de Linguagem Natural , Algoritmos , Humanos , SemânticaRESUMO
BACKGROUND: Medical trainees are required to learn many procedures following instructions to improve their skills. This study aims to investigate the pupillary response of trainees when they encounter moment of performance difficulty (MPD) during skill learning. Detecting the moment of performance difficulty is essential for educators to assist trainees when they need it. METHODS: Eye motions were recorded while trainees practiced the thoracostomy procedure in the simulation model. To make pupillary data comparable among trainees, we proposed the adjusted pupil size (APS) normalizing pupil dilation for each trainee in their entire procedure. APS variables including APS, maxAPS, minAPS, meanAPS, medianAPS, and max interval indices were compared between easy and difficult subtasks; the APSs were compared among the three different performance situations, the moment of normal performance (MNP), MPD, and moment of seeking help (MSH). RESULTS: The mixed ANOVA revealed that the adjusted pupil size variables, such as the maxAPS, the minAPS, the meanAPS, and the medianAPS, had significant differences between performance situations. Compared to MPD and MNP, pupil size was reduced during MSH. Trainees displayed a smaller accumulative frequency of APS during difficult subtask when compared to easy subtasks. CONCLUSIONS: Results from this project suggest that pupil responses can be a good behavioral indicator. This study is a part of our research aiming to create an artificial intelligent system for medical trainees with automatic detection of their performance difficulty and delivering instructional messages using augmented reality technology.
Assuntos
Pupila , Simulação por Computador , Humanos , Pupila/fisiologiaRESUMO
INTRODUCTION: We evaluated the velocity profiles of patients with lateral collateral ligament (LCL) injuries of the ankle with a goal of understanding the control mechanism involved in walking. METHODS: We tracked motions of patients' legs and feet in 30 gait cycles recorded from patients with LCL injuries of the ankle and compared them to 50 gait cycles taken from normal control subjects. Seventeen markers were placed on the foot following the Heidelberg foot measurement model. Velocity profiles and microadjustments of the knee, ankle, and foot were calculated during different gait phases and compared between the patient and control groups. RESULTS: Patients had a smaller first rocker percentage and larger second rocker percentage in the gait cycle compared to controls. Patients also displayed shorter stride length and slower strides and performed more microadjustments in the second rocker phase than in other rocker/swing phases. Patients' mean velocities of the knee, ankle, and foot in the second rocker phase were also significantly higher than that in control subjects. Discussion. Evidence from velocity profiles suggested that patients with ligament injury necessitated more musculoskeletal microadjustments to maintain body balance, but these may also be due to secondary injury. Precise descriptions of the spatiotemporal gait characteristics are therefore crucial for our understanding of movement control during locomotion.
RESUMO
In recent years, convolutional neural network (CNN)-based object detection algorithms have made breakthroughs, and much of the research corresponds to hardware accelerator designs. Although many previous works have proposed efficient FPGA designs for one-stage detectors such as Yolo, there are still few accelerator designs for faster regions with CNN features (Faster R-CNN) algorithms. Moreover, CNN's inherently high computational complexity and high memory complexity bring challenges to the design of efficient accelerators. This paper proposes a software-hardware co-design scheme based on OpenCL to implement a Faster R-CNN object detection algorithm on FPGA. First, we design an efficient, deep pipelined FPGA hardware accelerator that can implement Faster R-CNN algorithms for different backbone networks. Then, an optimized hardware-aware software algorithm was proposed, including fixed-point quantization, layer fusion, and a multi-batch Regions of interest (RoIs) detector. Finally, we present an end-to-end design space exploration scheme to comprehensively evaluate the performance and resource utilization of the proposed accelerator. Experimental results show that the proposed design achieves a peak throughput of 846.9 GOP/s at the working frequency of 172 MHz. Compared with the state-of-the-art Faster R-CNN accelerator and the one-stage YOLO accelerator, our method achieves 10× and 2.1× inference throughput improvements, respectively.
RESUMO
Training models to predict click and order targets at the same time. For better user satisfaction and business effectiveness, multitask learning is one of the most important methods in e-commerce. Some existing researches model user representation based on historical behaviour sequence to capture user interests. It is often the case that user interests may change from their past routines. However, multi-perspective attention has broad horizon, which covers different characteristics of human reasoning, emotions, perception, attention, and memory. In this paper, we attempt to introduce the multi-perspective attention and sequence behaviour into multitask learning. Our proposed method offers better understanding of user interest and decision. To achieve more flexible parameter sharing and maintaining the special feature advantage of each task, we improve the attention mechanism at the view of expert interactive. To the best of our knowledge, we firstly propose the implicit interaction mode, the explicit hard interaction mode, the explicit soft interaction mode, and the data fusion mode in multitask learning. We do experiments on public data and lab medical data. The results show that our model consistently achieves remarkable improvements to the state-of-the-art method.
Assuntos
Aprendizagem , Aprendizado de Máquina , HumanosRESUMO
The satisfaction of employees is very important for any organization to make sufficient progress in production and to achieve its goals. Organizations try to keep their employees satisfied by making their policies according to employees' demands which help to create a good environment for the collective. For this reason, it is beneficial for organizations to perform staff satisfaction surveys to be analyzed, allowing them to gauge the levels of satisfaction among employees. Sentiment analysis is an approach that can assist in this regard as it categorizes sentiments of reviews into positive and negative results. In this study, we perform experiments for the world's big six companies and classify their employees' reviews based on their sentiments. For this, we proposed an approach using lexicon-based and machine learning based techniques. Firstly, we extracted the sentiments of employees from text reviews and labeled the dataset as positive and negative using TextBlob. Then we proposed a hybrid/voting model named Regression Vector-Stochastic Gradient Descent Classifier (RV-SGDC) for sentiment classification. RV-SGDC is a combination of logistic regression, support vector machines, and stochastic gradient descent. We combined these models under a majority voting criteria. We also used other machine learning models in the performance comparison of RV-SGDC. Further, three feature extraction techniques: term frequency-inverse document frequency (TF-IDF), bag of words, and global vectors are used to train learning models. We evaluated the performance of all models in terms of accuracy, precision, recall, and F1 score. The results revealed that RV-SGDC outperforms with a 0.97 accuracy score using the TF-IDF feature due to its hybrid architecture.
RESUMO
In the rapid development of various technologies at the present stage, representative artificial intelligence technology has developed more prominently. Therefore, it has been widely applied in various social service areas. The application of artificial intelligence technology in tax consultation can optimize the application scenarios and update the application mode, thus further improving the efficiency and quality of tax data inquiry. In this paper, we propose a novel model, named RDN-MESIM, for paraphrase identification tasks in the tax consulting area. The main contribution of this work is designing the RNN-Dense network and modifying the original ESIM to adapt to the RDN structure. The results demonstrate that RDN-MESIM obtained a better performance as compared to other existing relevant models and archived the highest accuracy, of up to 97.63%.
Assuntos
Inteligência ArtificialRESUMO
BACKGROUND: Traditional Chinese medicine (TCM) clinical records contain the symptoms of patients, diagnoses, and subsequent treatment of doctors. These records are important resources for research and analysis of TCM diagnosis knowledge. However, most of TCM clinical records are unstructured text. Therefore, a method to automatically extract medical entities from TCM clinical records is indispensable. OBJECTIVE: Training a medical entity extracting model needs a large number of annotated corpus. The cost of annotated corpus is very high and there is a lack of gold-standard data sets for supervised learning methods. Therefore, we utilized distantly supervised named entity recognition (NER) to respond to the challenge. METHODS: We propose a span-level distantly supervised NER approach to extract TCM medical entity. It utilizes the pretrained language model and a simple multilayer neural network as classifier to detect and classify entity. We also designed a negative sampling strategy for the span-level model. The strategy randomly selects negative samples in every epoch and filters the possible false-negative samples periodically. It reduces the bad influence from the false-negative samples. RESULTS: We compare our methods with other baseline methods to illustrate the effectiveness of our method on a gold-standard data set. The F1 score of our method is 77.34 and it remarkably outperforms the other baselines. CONCLUSIONS: We developed a distantly supervised NER approach to extract medical entity from TCM clinical records. We estimated our approach on a TCM clinical record data set. Our experimental results indicate that the proposed approach achieves a better performance than other baselines.
RESUMO
Rhizosphere microbial communities are known to be related to plant health; using such an association for crop management requires a better understanding of this relationship. We investigated rhizosphere microbiomes associated with Verticillium wilt symptoms in two cotton cultivars. Microbial communities were profiled by amplicon sequencing, with the total bacterial and fungal DNA quantified by quantitative polymerase chain reaction based on the respective 16S and internal transcribed spacer primers. Although the level of V. dahliae inoculum was higher in the rhizosphere of diseased plants than in the healthy plants, such a difference explained only a small proportion of variation in wilt severities. Compared to healthy plants, the diseased plants had much higher total fungal/bacterial biomass ratio, as represented by quantified total fungal or bacterial DNA. The variability in the fungal/bacterial biomass ratio was much smaller than variability in either fungal or bacterial total biomass among samples within diseased or healthy plants. Diseased plants generally had lower bacterial alpha diversity in their rhizosphere, but such differences in the fungal alpha diversity depended on cultivars. There were large differences in both fungal and bacterial communities between diseased and healthy plants. Many rhizosphere microbial groups differed in their abundance between healthy and diseased plants. There was a decrease in arbuscular mycorrhizal fungi and an increase in several plant pathogen and saprophyte guilds in diseased plants. These findings suggested that V. dahliae infection of roots led to considerable changes in rhizosphere microbial communities, with large increases in saprophytic fungi and reduction in bacterial community.
RESUMO
Particle size is the most important index to reflect the crushing quality of ores, and the accuracy of particle size statistics directly affects the subsequent operation of mines. Accurate ore image segmentation is an important prerequisite to ensure the reliability of particle size statistics. However, given the diversity of the size and shape of ores, the influence of dust and light, the complex texture and shadows on the ore surface, and especially the adhesion between ores, it is difficult to segment ore images accurately, and under-segmentation can be a serious problem. The construction of a large, labeled dataset for complex and unclear conveyor belt ore images is also difficult. In response to these challenges, we propose a novel, multi-task learning network based on U-Net for ore image segmentation. To solve the problem of limited available training datasets and to improve the feature extraction ability of the model, an improved encoder based on Resnet18 is proposed. Different from the original U-Net, our model decoder includes a boundary subnetwork for boundary detection and a mask subnetwork for mask segmentation, and information of the two subnetworks is fused in a boundary mask fusion block (BMFB). The experimental results showed that the pixel accuracy, Intersection over Union (IOU) for the ore mask (IOU_M), IOU for the ore boundary (IOU_B), and error of the average statistical ore particle size (ASE) rate of our proposed model on the testing dataset were 92.07%, 86.95%, 52.32%, and 20.38%, respectively. Compared to the benchmark U-Net, the improvements were 0.65%, 1.01%, 5.78%, and 12.11% (down), respectively.
RESUMO
Traditional Chinese medicine (TCM) symptom normalization is difficult because the challenges of the symptoms having different literal descriptions, one-to-many symptom descriptions and different symptoms sharing a similar literal description. We propose a novel two-step approach utilizing hierarchical semantic information that represents the functional characteristics of symptoms and develop a text matching model that integrates hierarchical semantic information with an attention mechanism to solve these problems. In this study, we constructed a symptom normalization dataset and a TCM normalization symptom dictionary containing normalization symptom words, and assigned symptoms into 24 classes of functional characteristics. First, we built a multi-label text classifier to isolate the hierarchical semantic information from each symptom description and count the corresponding normalization symptoms and filter the candidate set. Then we designed a text matching model of mixed multi-granularity language features with an attention mechanism that utilizes the hierarchical semantic information to calculate the matching score between the symptom description and the normalization symptom words. We compared our approach with other baselines on real-world data. Our approach gives the best performance with a Hit@ 1, 3, and 10 of 0.821, 0.953, and 0.993, respectively, and a MeanRank of 1.596, thus outperforming significantly regarding the symptom normalization task. We developed an approach for the TCM symptom normalization task and demonstrated its superior performance compared with other baselines, indicating the promise of this research direction.
Assuntos
Semântica , Envio de Mensagens de Texto , Idioma , Medicina Tradicional Chinesa , Processamento de Linguagem NaturalRESUMO
In recent years, image processing methods based on convolutional neural networks (CNNs) have achieved very good results. At the same time, many branch techniques have been proposed to improve accuracy. Aiming at the change detection task of remote sensing images, we propose a new network based on U-Net in this paper. The attention mechanism is cleverly applied in the change detection task, and the data-dependent upsampling (DUpsampling) method is used at the same time, so that the network shows improvement in accuracy, and the calculation amount is greatly reduced. The experimental results show that, in the two-phase images of Yinchuan City, the proposed network has a better antinoise ability and can avoid false detection to a certain extent.
Assuntos
Redes Neurais de Computação , Tecnologia de Sensoriamento Remoto , Processamento de Imagem Assistida por ComputadorRESUMO
Recent advances in convolutional neural networks (CNNs) have shown impressive results in semantic segmentation. Among the successful CNN-based methods, U-Net has achieved exciting performance. In this paper, we proposed a novel network architecture based on U-Net and atrous spatial pyramid pooling (ASPP) to deal with the road extraction task in the remote sensing field. On the one hand, U-Net structure can effectively extract valuable features; on the other hand, ASPP is able to utilize multiscale context information in remote sensing images. Compared to the baseline, this proposed model has improved the pixelwise mean Intersection over Union (mIoU) of 3 points. Experimental results show that the proposed network architecture can deal with different types of road surface extraction tasks under various terrains in Yinchuan city, solve the road connectivity problem to some extent, and has certain tolerance to shadows and occlusion.
Assuntos
Insuficiência Adrenal , Retardo do Crescimento Fetal , Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Osteocondrodisplasias , Tecnologia de Sensoriamento Remoto , Anormalidades Urogenitais , Algoritmos , Análise de Dados , Humanos , Processamento de Imagem Assistida por Computador/métodosRESUMO
A heterogeneous method of coupled multiscale strength model is presented in this paper for calculating the strength of medical polyesters such as polylactide (PLA), polyglycolide (PGA) and their copolymers during degradation by bulk erosion. The macroscopic device is discretized into an array of mesoscopic cells. A polymer chain is assumed to stay in one cell. With the polymer chain scission, it is found that the molecular weight, chain recrystallization induced by polymer chain scissions, and the cavities formation due to polymer cell collapse play different roles in the composition of mechanical strength of the polymer. Therefore, three types of strength phases were proposed to display the heterogeneous strength structures and to represent different strength contribution to polymers, which are amorphous phase, crystallinity phase and strength vacancy phase, respectively. The strength of the amorphous phase is related to the molecular weight; strength of the crystallinity phase is related to molecular weight and degree of crystallization; and the strength vacancy phase has negligible strength. The vacancy strength phase includes not only the cells with cavity status but also those with an amorphous status, but a molecular weight value below a threshold molecular weight. This heterogeneous strength model is coupled with micro chain scission, chain recrystallization and a macro oligomer diffusion equation to form a multiscale strength model which can simulate the strength phase evolution, cells status evolution, molecular weight, degree of crystallinity, weight loss and device strength during degradation. Different example cases are used to verify this model. The results demonstrate a good fit to experimental data.