This study focuses on the reasoning of symptoms of TCM internal diseases. Taking cough as an example, the BERT-BiLSTM-CRF model is used for entity recognition. Experiments show that the model has the best entity recognition effect on three types of texts, including teaching cases, clinical cases and literature data, and the F1 value is up to 0.967. It can effectively identify symptoms, course of disease, tongue condition and pulse condition in TCM diagnosis and treatment texts, which lays a solid foundation for intelligent assisted syndrome differentiation and treatment of TCM. By constructing four types of entity corpora and using three mainstream entity recognition models for experiments, it is found that the BERT-BiLSTM-CRF model has a good entity recognition effect in three types of data: teaching cases, clinical cases and literature data. Experimental results show that the F1 values of the BERT-bilSTM-CRF model in teaching cases, clinical cases and literature data are 0.967, 0.82 and 0.91, respectively, which provides an effective method for information extraction in the field of TCM syndrome differentiation diagnosis and treatment, and lays a foundation for further research on knowledge reasoning.

A field experiment was conducted for 2 years in Green Tea Laboratory of Saitama Prefectural Agriculture and Forestry Research Center, Iruma, Saitama, Japan from March 2014 to December 2015. Controlled release fertilizers (CRF) or organic fertilizers (ORG) which is the mixture of chicken manure and oil cakewere applied with the amount of 450 kg N ha^-1 yr^-1 in 2014 and 397 kg N ha^-1 yr^-1 in 2015. Nitrous oxide (N₂O) emissionsfrom soil in green tea fields were measured by closed chamber method. The results showed that CRF has significantly lower N₂O compared to ORG. The cumulative N₂O emissions from CRF accounted for 51% of N₂O emissions from ORG fields and 138% of control with no fertilizer treatment. The N₂O flux from the row was higher than that of under the canopy, since fertilizer were applied on the row. However the total emission from the area between the rows was lower than that under the canopy because of the area ratio of row and canopy was 1:5.

Accurate assessment of the breast cancer deterioration degree plays a crucial role in making medical plan, and the important basis for degree assessment is the number of mitoses in a given area of the pathological image. We utilized deep multi-scale fused fully convolutional neural network (MFF-CNN) combing with conditional random felid (CRF) to detect mitoses in hematoxylin and eosin stained histology image. Analyze the characteristics of mitotic detection ----scale invariance and sparsity, as well as the difficulties ---- small amount of data , inconsistent image staining and sample class unbalanced. Based on this, mitotic detection model is designed. In this paper, a tissue-based staining equalization method is used, and to establish an effective training sample set, we select training samples by using CNN. A mitotic detection model fusing multi-level and multi-scale features and context information was designed, and the corresponding training strategy was made to reduce over-fitting. As preliminarily validated on the public 2014 ICPR MITOSIS data, our method achieves a better performance in term of detection accuracy than ever recorded for this dataset.

When analyzing user geospatial information through deep learning methods, it is typically necessary to annotate existing geospatial data. Currently, manual annotation methods are commonly employed, suffering from issues related to low efficiency and accuracy. This design is based on the TensorFlow deep learning framework and first realizes the BiLSTM+CRF+AGG deep learning model. Among the models, AGG is the aggregation layer, which is introduced to solve the problem of solid particle size equilibrium. Secondly, based on the characteristics of original data and professional data, an automatic labeling algorithm is proposed. The algorithm first preprocesses the acquired original data set and professional data set, The most valuable unlabeled data set which can make the training model converge quickly is selected from the original data set as the target data set. Sort the target data set based on preset rules and set annotation parameters for the sorted target data set. Based on the set annotation parameters, the corpus is synthesized and used as the annotation result. Thirdly, based on the active learning strategy, a manual annotation auxiliary scheme is proposed, and an Excel generation module that is convenient for manual annotation correction is designed and developed to further improve the efficiency and quality of annotation through iterative processing. The combination of the BiLSTM+CRF+AGG deep learning model and high-quality annotation can accurately identify non-standard, incomplete, or even incorrect geographical information entities. This annotation method has found successful practical application within the context of the research project and has been granted an invention patent.

Geospatial knowledge in massive academic papers can provide knowledge services such as location-based research hotspot analysis, spatio-temporal data aggregation, research results recommendation, etc. However, geospatial knowledge often exists implicitly in literature resources in unstructured form, which is difficult to be directly accessed and mined and utilized for rapid production of massive thematic maps. In this paper, we take the geospatial knowledge of the area studied as an example and introduce its extraction method in detail. An integrated feature template matching and random forest classification algorithm is proposed for accurately identifying research areas from the abstract texts of academic papers and producing thematic maps. Firstly, the precise recognition of geographical names is achieved step by step based on BiLSTM-CRF algorithm and improved heuristic disambiguation method; then, the area studied is extracted by the designed integrated feature recognition template of area studied using random forest classification algorithm, and a fast thematic map is designed for the knowledge of area studied, topic and literature. The experimental results show that the area studied recognition accuracy can reach 97%, the F-value is 96%, and the recall rate reaches 96%, achieving high accuracy and high efficiency of area studied extraction in text. Based on the geospatial knowledge, the thematic map can achieve the effect of fast map formation and accurate expression.

The interpretation of land use and land cover (LULC) is an important issue in the fields of high-resolution remote sensing (RS) image processing and land resource management. Fully training a new or existing convolutional neural network (CNN) architecture for LULC classification requires a large amount of remote sensing images. Thus, fine-tuning a pre-trained CNN for LULC detection is required. To improve the classification accuracy for high resolution remote sensing images, it is necessary to use another feature descriptor and to adopt a classifier for post-processing. A fully connected conditional random fields (FC-CRF), to use the fine-tuned CNN layers, spectral features, and fully connected pairwise potentials, is proposed for image classification of high-resolution remote sensing images. First, an existing CNN model is adopted, and the parameters of CNN are fine-tuned by training datasets. Then, the probabilities of image pixels belong to each class type are calculated. Second, we consider the spectral features and digital surface model (DSM) and combined with a support vector machine (SVM) classifier, the probabilities belong to each LULC class type are determined. Combined with the probabilities achieved by the fine-tuned CNN, new feature descriptors are built. Finally, FC-CRF are introduced to produce the classification results, whereas the unary potentials are achieved by the new feature descriptors and SVM classifier, and the pairwise potentials are achieved by the three-band RS imagery and DSM. Experimental results show that the proposed classification scheme achieves good performance when the total accuracy is about 85%.

Proteins are the workhorses of the cell that perform biological functions by interacting with other proteins. Many statistical methods for protein-protein interaction (PPI) have been studied without considering time-dependent changes in networks and the functionalities. I introduced a novel method that models PPI networks as being dynamic in nature and evolving time-varying multivariate distribution with Conditional Random Fields (CRF). This research is directed towards implementing this new combinatorial algorithm on massively parallel architectures such as Graphics Processing Units (GPUs) for efficient computations for large scale bioinformatics datasets. I compared Conditional Random Fields (CRF) and the proposed novel method using CRF combined with the Block Coordinate Descent algorithm for human protein-protein interaction data set. Both are implemented on GPU-Accelerated Computing Architecture and the proposed novel method showed the advantages in predicting protein-protein interaction sites. I also show that the proposed approach is more efficient in 6.13% than standalone CRF++ in predicting protein-protein interaction sites.

This review focuses on the emerging evidence for the association between estimated cardiorespiratory fitness (eCRF) and metabolic risk factors. Given the challenges associated with directly measuring cardiorespiratory fitness (CRF) in large populations, eCRF presents a practical alternative for predicting metabolic health risks. A literature search identified six relevant cohort studies from 2021 to 2024 that investigated the association of eCRF with hypertension, hyperglycemia, dyslipidemia, and obesity. The review findings highlight a consistent inverse relationship between higher eCRF and lower incidence of metabolic risks in line with CRF cohort studies. This review highlights the importance of low eCRF as a primordial indicator for metabolic risks and underscores the need for broader application. Future research directions should include exploring eCRF's predictive accuracy across diverse populations and health outcomes and testing its real-world applicability in healthcare and public health settings.

Cemented rock fill (CRF) is commonly used in cut-and-fill stoping operation in underground mining. This allows for the maximum recovery of ore. Backfilling can improve stope stability in underground workings, and then improve ground stability of the whole mine site. Backfilling step scenarios vary from site to site. This paper presents the investigation of five different backfilling step scenarios and their impacts on the stability of stopes at four different mining levels. A comprehensive comparison of displacements, major principal stress and stress concentration factor (SCF) was conducted. The results show that different backfilling step scenarios have little influence on the final displacement for displacement in the stopes. Among the five backfilling scenarios, the major principal stress and stress concentration factor (SCF) have almost the same final results. The backfilling scenario SCN-1 is the optimum option among these five backfilling scenarios. It can immediately prevent the increase of the displacement and reduce the sidewall stress concentration, thereby preventing possible failures. Using the same strength of CRF can achieve same effects among the four mining levels. Applying backfilling CRF of the same strength at different mining depths is acceptable and feasible to improve the stability of the stopes.

Alcohol use disorder (AUD) is a chronically relapsing disease characterized by loss of control in seeking and consuming alcohol (ethanol) driven by recruitment of brain stress systems. However, AUD differs among the sexes: men are more likely to develop AUD, but women progress from casual to binge drinking and heavy alcohol use more quickly. The central amygdala (CeA) is a hub of stress and anxiety, with corticotropin releasing factor (CRF)-CRF1 receptor and GABAergic signaling dysregulation occurring in alcohol dependent male rodents. However, we recently showed that GABAergic synapses in female rats are less sensitive to the acute effects of ethanol. Here, we used patch clamp electrophysiology to examine the effects of alcohol dependence on the CRF-modulation of rat CeA GABAergic transmission of both sexes. We found that GABAergic synapses of naïve female rats were unresponsive to CRF application compared males, although alcohol dependence induced a similar CRF responsivity in both sexes. In situ hybridization revealed that females had less CeA neurons containing mRNA for the CRF1 receptor (Crhr1) than males, but in dependence, the percentage of Crhr1-expressing neurons in females increased, unlike males. Overall, our data provide evidence for sexually dimorphic CeA CRF system effects on GABAergic synapses in dependence.