Abstract:Motivated by fractional differential equations, we consider the well-posedness of a class of singular Volterra integral equations on the L^p (p ≥1) space by using the fixed point theorem, generalize and improve the known results. Particularly, the well-posedness of Riemann--Liouville fractional order ordinary differential equations can be regarded as a special case of our results.

Abstract:In this paper, we investigate the maximum likelihood estimator of the parameter of a partially observable coupled stochastic parabolic equation driven by additive white Gaussian noises in time and space. For fixed observation time and noise intensity, the estimator is proved to be asymptotically consistent and with asymptotic normality. A numerical example is provided to illustrate the theoretic results.

Abstract:We in this paper study the existence of solutions for a class of elastic beam problems with simply supported ends. The proof of the main results is based on the method of lower and upper solutions and the Elias's inequality.

Abstract:In this paper, we study the existence and uniqueness of positive solutions for the periodic boundary value problems of a class of first order ordinary differential equations. The proof of the main results is based on the Schauder fixed point theorem and the Leray-Schauder degree theory.

Abstract:In this paper, a three-level linear finite difference scheme with high theoretical accuracy is proposed for the initial-boundary value problem of Rosenau-KdV equation. This scheme simulates two conservative properties very well. The existence and uniqueness of the difference solution and prior estimates are obtained. Then the convergence and stability of the scheme are analyzed by using the energy method. Numerical examples verify the theoretical results.

Abstract:Aiming at the problems of poor contrast and poor visual quality in dark area of low illumination image, this paper proposes a new low illumination image enhancement algorithm. This method can effectively avoid the problems of color distortion and over enhancement of bright areas caused by traditional enhancement methods. Firstly , the authors design a linear mapping function to compress the gray area with few pixels, which can improve the overall brightness of the image while keeping the gray distribution characteristics of the image unchanged. Then, the authors use the gray distribution of the image to generate the corresponding enhancement mapping function. Finally, the authors use image fusion technology to further improve the overall color performance of the image. Experimental results show that the proposed algorithm has better performance in terms of color performance and brightness enhancement, and is better than the most advanced methods in visual quality and quantitative measurement.

Abstract:Drug-Drug interaction refers to the mutual promotion or inhibition between drugs. For the existing drug relationship extraction methods, the use of external background knowledge and natural language processing tools leads to the problem of error propagation and accumulation, and most existing studies blind drug entities at the data preprocessing stage, ignoring the target drug entity information that is helpful to identify the relationship category. In this paper, a drug interaction extraction model based on pretrained biomedical language model and word map neural network is proposed. In this model, the original feature representation of sentences is obtained by pretrained language model, and the global feature information representation of sentences is obtained by convolution operation on the word map constructed based on data set. Finally, the feature representation of drug interaction relationship extraction task was constructed by stitching the feature with drug target entities, which can not only obtain rich global feature information but also avoid using natural language processing tools and external background knowledge, and improve the accuracy of the model. The F1 value of the model on the DDIExtraction 2013 dataset achieved 83.25%, which outperforms the current latest methods by 2.35%.

Abstract:The task of named entity recognition is to locate the entities in the text and classify them into predefined categories. The current mainstream Chinese named entity recognition models are characterbased named entity recognition models which word segmentation is required before using syntactic features， syntactic information of sentences cannot be well utilized as a result. In addition， the characterbased models cannot make use of the prior dictionary information and the pictographic information contained in Chinese radicals. To solve the above problems， this paper proposes a multifeature Chinese named entity recognition model combining syntax and multigranularity semantic information. The experiments demonstrate that the proposed model is better than the current mainstream Chinese named entity recognition models， the influence of various features on the Chinese entity recognition effect is analyzed through experiments as well.

Abstract:In the field of data scarcity, the performance of named entity recognition is limited by the expression of underfitting word features. The named entity recognition effect can be improved by introducing conventional multitask learning methods, but additional labeling costs are required. Aiming at addressing this problem, we propose a new named entity recognition method based on multigranularity cognition, which can enhance the character feature information and improve the performance of named entity recognition without incurring additional tagging costs. In order to optimize the expression of word embedding, in this approach, we start from the multi granularity cognition theory and use BiLSTM and CRF as the basic model, the task of named entity recognition under word granularity is combined with the task of entity number prediction under sentence global granularity. Multiple experiments on three different types of data sets show that the method of introducing multigranularity cognition method can effectively improve the performance of named entity recognition.

Abstract:Generating digital mid-fidelity prototypes from hand-drawn sketches is a necessary step in User Interface design, which requires a lot of time and energy for designers. In order to solve the problem, this paper proposes an automatic method for generating mid-fidelity prototypes based on object detection and heuristic rules, and implements an automatic tool based on this method. Firstly, an object detection model SA-FRCNN based on the Shuffle Attention (SA) mechanism is proposed to detect components in hand-drawn sketches; and then, heuristic rules are proposed to optimize the layout of the detected results. Experiments show that our method can improve the detection accuracy of hand-drawn sketches by 2.3% compared with the baseline model; in addition, our tool can effectively improve the efficiency of prototyping.

Abstract:In the field of network security, the threat of malicious code is an unavoidable topic. How to quickly detect malicious code, prevent and reduce the harm caused by malicious code has always been an urgent problem. This paper proposes a malicious code detection method based on the behavior relation network. First, obtain the behavior report by executing the sample in the sandbox, and then construct a behavior relationship network by extracting the three behavior records of the sample''s API call, registry access, and file read and write operations from the behavior report . The constructed behavior relationship network includes "PE", "API", "Registry" and "File" 4 types of nodes, we then use a metagraphbased method to calculate the similarity matrix between samples, and finally the Support Vector Machine （SVM） model, which kernel is custom defined, is used for training and prediction. Experimental results show that the method proposed in this paper can achieve a classification accuracy of 95.5% and can effectively detect malicious code.

Abstract:The combination of blockchain technology and internet of things can play a decentralized advantage and improve the data security and reliability of the internet of things system to a certain extent, but the flux limitation characteristics of the blockchain make the low transaction throughput of blockchain difficult to meet the high throughput business requirements in the internet of things scenario. In this paper, a minimum loss function algorithm is proposed，in which the state space and behavior space are first constructed according to the state and action input, and then the behavior value function of state space and behavior space are calculated iteratively under the condition of system delay constraint. Finally, the block size and block interval is adjusted by using the loss function to compare the true value and estimated value of the behavior value function. The simulation results show that，compared with the DDRL algorithm and the DRL algorithm, the minimum loss algorithm dynamically adjusts the block size and block spacing, and can obtain higher throughput after the blockchain based internet of things system reaches stablility.

Abstract:In order to address the drawbacks such as a large amount of calculation, the difficulty in balancing convergence speed, and uniformity of population distribution when solving multiobjective optimization problems with traditional evolutionary algorithms, a leading NSGAII algorithm based on regional unbalanced subspace (NSGAII URS) is proposed. First, based on the NSGAII algorithm and the local search algorithm, the population leading solution set is added in each genetic process to guide the population to converge quickly. Then the target space, where the nondominated solution is located, is evenly divided, the concepts of sparse subspace and free subspace are introduced. Finally, the unbalanced subspace is optimized by a local search strategy based on sparse degree to further improve the uniformity of the population distribution. The proposed algorithm is compared with five other advanced multiobjective evolutionary algorithms, verified by benchmark test function, and two general indicators of inverse generation distance (IGD) and hypervolume (HV) are used for performance evaluation. Experimental results show that the proposed NSGAII algorithm is significantly better than other compared multiobjective optimization algorithms in terms of solution distribution and convergence.

Abstract:Tactics, Techniques, and Procedures (TTPs) analysis in Cyber Threat Intelligence (CTI) providing a global view of cyberattack events and reveal system weaknesses, is a key technique for cyberattack traceability. Existing TTPs classification schemes are poorly and unevenly oriented to abstract language environments. In this paper, we propose a multi-label deep learning model based on association enhancement: RENet, which classifies tactics and techniques by using a multi-label classifier that combines contextual information and multiple word meanings, and enhances technique classification by transferring the classification results of the original tactics through a conditional transfer matrix from tactics to techniques. Experiments show that RENet has more accurate classification results of tactics and techniques with faster convergence than other classification models. The F1 scores of RENet for techniques and tactics classification are 4.62% and 0.78% higher than the best existing models on the English dataset, and 3.95% and 3.77% higher on the Chinese dataset, respectively.

Abstract:Ancient literature of Traditional Chinese Medicine (TCM) contains rich clinical experiences, which is the empirical summary of clinical diagnosis and treatment in the process of ancient Chinese medicine practice, and embodies the theoretical framework and ideological basis of the formation and development of TCM. However, due to the volume and dispersion of valuable clinical experiences, it is difficult for TCM doctors to quickly and comprehensively obtain the clinical information they need from ancient literature manually, and the document retrieval tools can only provide documentlevel information screening, which cannot support finegrained information extraction. In addition, the different characteristics of ancient Chinese relative to modern Chinese also limit the use of mainstream text analysis tools. For this reason, we propose a task of information extraction from the ancient literature of TCM for obtaining clinical experiences, which is used to identify text fragments describing clinical experiences in ancient literature and manually annotate sample data for training and testing the extraction task, a sequence labeling model is designed based on deep learning to complete the task. Considering the overfitting problem that can be brought about by the small amount of annotated data, we introduce adversarial training and virtual adversarial training to enhance the generalization ability of the proposed model. A series of sufficient experiments are conducted on the clinical experience dataset to verify the effectiveness of the model, and the experimental results show the feasibility of extracting clinical experiences from ancient literature by information extraction technology, and a promising baseline and a reusable annotated dataset for the new information extraction task are available.

Abstract:To analyze the generalized thermoelastic problem in two-dimensional composite material, a new two-dimensional vertex-center finite volume method (CV-FVM) has been developed based on Lord-Shulman (L-S), Green-Lindsay (G-L) and traditional coupled theories. Using the staggered grid technique, the unknown variable is defined at the cell vertex, while the material property is defined at the cell center. The space terms of governing equations are discretized by bilinear quadrilateral element, and the time terms are discretized by Euler implicit formula. Thermal shock problem in infinite plate with homogeneous material is studied by CV-FVM. The results show that the present method can effectively capture the temperature jump and thermoelastic coupling characteristics at the front of the thermal wave and elastic wave. Then, the developed CV-FVM was used to study the thermal shock problem in composite with Ti-6Al-4V/ZrO2 with different material constant p, the results show that the value p=1 minimizes the maximum (tensile) stress applied at the middle of the functionally graded layer under L-S theory, and the value p=10 minimizes the maximum (tensile) stress under G-L and T-C theories. The effects of p on interfacial thermoelastic response is different under different coupling theories, one cannot conclude that a linear variation of the properties minimizes the maximum stress. The developed method can be used as an alternative tool for solving thermal wave and generalized thermoelastic problems.

Abstract:Jet impingement is a heat exchange method with high local heat exchange efficiency, and it is very important for engineering application. In this study, by means of fluid simulation software Fluent, a jet impingement cooling model with multiple nozzles was designed, the steady-state heat transfer characteristics of the process was studied when the combined jet impinges on the wall surface vertically and obliquely, and the influence of nozzle inclination angle and distance on the heat transfer characteristics of the wall was discussed. It is found that as the inclination angle of the oblique nozzle increases, the average Nusselt number of the combined jet gradually increases and then decreases. The combined jet inherits the advantages of the single straight jet and the oblique jet. It ensures heat transfer efficiency in the stagnation zone, improves the heat transfer efficiency downstream of the jet effectively and makes the wall temperature distribution more uniform in the meantime. When the oblique nozzle is close to the straight nozzle, the overall heat transfer characteristics of the combined jet are similar to that of the single oblique jet. When the distance between the inclined nozzle and the straight one is increased along the horizontal direction and the longitudinal direction, the average Nusselt number of the wall increases, the two high-temperature regions upstream from the nozzle area shift to the downstream, and the cooling efficiency improves significantly.

Abstract:In order to improve the plasma density and the ionization rate of working medium gas, the helical wave generated by helical antenna is used to excite Ar plasma, and the characteristics of ion density and electron temperature of plasma are analyzed by RF compensated Langmuir probe. The experimental results show that with the increase of air pressure and power, the helicon plasma enters the helicon discharge mode in advance. Under the pressure of 1.0 Pa, when the RF power reaches 400 W, the plasma discharge enters the helicon discharge mode, and the plasma density in the extended region exceeds 1e18 m-3. The electron density is the highest in the center of the discharge tube and gradually decreases along the radial direction. The research results will provide the basis and experience for the large volume H2 helicon plasma.

Abstract:The scintillation chamber with high detection efficiency is the main method for radon measurement, but it needs to stand for 3h after sampling before starting measurement. In order to meet the needs of rapid measurement of radon in scintilators, a rapid measurement method for radon in scintilators was established according to the decay relationship of radon and its progeny under the same efficiency condition of Rn-222, Po-218 and Bi-214. Based on the integral counting method, the calibration factor K0 at the measurement period of 60min was used to calculate the theoretical calibration factors at the measurement period of 30 min and 15 min. A 54 mL scintillation chamber with a detection efficiency of approximately 100% for radon and its daughters was selected to carry out rapid calibration experiments at a radon concentration of 20 Bq/mL. The experimental results show that the error between the experimental value and the theoretical value is within 5%, which can realize the rapid measurement of radon in scintillation chamber.

Abstract:In the structures of the illite and montmorillonite, because the amount of Al3+ replacing Si4+ in the tetrahedrons is different, the number of layer charge is different, and the degrees of damage to the structure during acid treatment are also different. In this work, two raw materials, Zhejiang Anji bentonite and Jilin Antu illite, were respectively acid treated with sulfuric acid and hydrogen peroxide for comparative experimental study. The phases, structures, thermal properties, spectroscopic properties and microscopic morphologies of the raw were characterized and analyzed. The results show that the montmorillonite structure of the montmorillonite sample is destroyed when the acid treatment concentration is 2 mol/L. With the increase of the acid treatment concentration, the surface of the lamellae changes from smooth and flat to edge curling to collapse between layers; The structure and morphology of illite of the sample are basically intact when the acid treatment concentration is 6 mol/L, and the acid corrosion resistance of montmorillonite is not as good as that of illite. The results in this study are of significance to the application and development of illite and montmorillonite.

Abstract:Potassium sorbate is an acidic food preservative, which can effectively inhibit the activity of mold, yeast and aerobic bacteria, prolong the storage time of food, and maintain the original flavor of food. In this study, fluorescence nitrogen doped carbon quantum dots (NCQDs) with good water solubility was synthesized by one-step hydrothermal method. Based on the effective quenching of NCQDs by potassium sorbate, a fluorescence probe for rapid detection of potassium sorbate was constructed. Under the optimal experimental conditions, the concentration of potassium sorbate showed a good linear relationship with the fluorescence quenching intensity of NCQDs in the range of 3.0×10-5~ 1.0×10-4 mol/L and 1.0×10-7~3.0×10-5 mol/L, and the detection limit was 9.5×10-8 mol/L. It has been used for the determination of potassium sorbate in soda water and white vinegar with the recoveries of 98.25%~102.7% and 98.33%~101.8%.

Abstract:To investigate the way that the calcium-dependent protein kinase CPK11 involved in regulating ABA signal transduction, Yeast two hybrid (Y2H) assay and Bimolecular fluorescence complementation (BiFC) assay were used to analyze the relationship between CPK11 and ABA-responsisve element binding factors ABF4. Y2H assay showed that CPK11 and ABF4 had an interaction in vitro, and BIFC assay showed that CPK11 and ABF4 had an interaction in vivo. The above experiments together proved that there was a direct interaction between CPK11 and ABF4. As a homologous protein of CPK11, CPK4 also interacted with transcription factor ABF4 in plants. In summary, these results indicated that CPK11 and its homologous protein CPK4 may participate in calcium-mediated ABA signaling pathway by interacting with transcription factor ABF4.

Abstract:In order to explore the molecular mechanism of Agaricus bisporus cooling and fruiting, this experiment took Agaricus bisporus W192 as the object. High-throughput RNA sequencing technology was used to analyze the gene expression of mycelium treated with the ambient temperature lowered from 21.5 ℃ to 17.5 ℃ at a constant rate for 4 d, 6 d and 8 d. There were 1481 differentially expressed genes in 6 d, which were 6.9% and 34% higher than 8 d and 4 d, respectively. Functional clustering analysis of GO Ontology (GO Ontology) showed that differential genes were more distributed in cellular components and biological processes, of which 6 d was dominant in metabolic and cellular processes. In addition, the differential genes under stress response and fruit body differentiation items were obviously and basically up-regulated. However, no such differentially expressed genes appeared on 4 d and 8 d, and the differentially expressed genes on 6 d were more obviously involved in the metabolism of various amino acids and sugars in cells. Functional enrichment analysis of KEGG pathway showed that the differentially expressed genes were mainly enriched in amino acid and antibiotic biosynthesis pathways, of which 6 d was enriched in glycolysis and ribosomal biosynthesis pathways, while 4 d and 8 d did not. On 4 d and 6 d, the cooling stimulation was easy to produce high yield, but on 4 d, the fruiting was early, the density was high, and the stratification was not obvious which would affect the mushroom type. On 6 d, the yield and quality were more stable. This study revealed the expression patterns of mycelium under different cooling stimuli, which provided a scientific theory for environmental regulation of industrial Agaricus bisporus during bud stimulating stage.

Abstract:In order to study the chemical constituents of Toxicodendron vernicifluum leaves, silica gel column chromatography, C18 medium pressure column chromatography and high pressure preparative chromatography were used to separate and purify the ethyl acetate portion of the ethanol extract from Toxicodendron vernicifluum leaves, and the structures of the compounds were identified by spectrum data. Eight compounds were isolated and identified as: gallic acid (1)，1,3,6-tri-O-galloyl-β-D-glucose (2)，1,2,4,6-tetra-O-galloyl-β-D-glucose(3)，astragalin (4), 1,2,3,4,6-penta-O-galloyl-β-D-glucose (5)，β- sitosterol (6)，shikimic acid (7)，engeletin(8). Among them, compounds 2, 3, 4, 7 and 8 were isolated from this plant for the first time, and compounds 1, 5 and 6 were isolated from the leaves of the plant for the first time.

Abstract:Doxorubicin (DOX) is the most widely used anthracycline family chemotherapeutic drug in clinical practice. In order to study whether Visnagin (VIS) could alleviate DOX-induced hepatotoxicity and nephrotoxicity and its related mechaninsms, this study first constructed a DOX-induced mouse model with acute or chronic liver and kidney injury. Then the protective effect of VIS on liver and kidney injury was evaluated by observing the mortality of mouse liver and kidney cells and the changes of liver and kidney function. The TUNEL results showed that VIS can significantly reduced apoptosis of liver and cells in acute and chronic injury models, and VIS can also relieve liver and kindey damage in the chronic model, in which VIS reduced DOX-induced creatine / urea and ALT/AST levels. In addition, VIS significantly inhibit the formation of the TOP2-DNA covalent complex level induced by DOX, suggesting that VIS may have a protective effect by inhibiting the TOP2 pathway downstream of DOX. The results will shed light on the follow-up in-depth exploration of reducing the side effects of DOX.

Abstract:Pyrrosiapetiolosa was used to make ointment to study its antibacterial and wound healing effects at the animal level. The P.petiolosa was prepared into an ointment, and the rat skin excision trauma model was constructed and randomly divided into normal trauma group and Staphylococcus aureus（S.aureus） infection group. Each group was treated with natural recovery, P. petiolosa ointment and Jingwanhong ointment, and then the weight and eating status of rats, wound healing, wound tissue pathology and serum inflammatory factor expression were measured. The results showed that the body weight and food intake of rats in different treatment groups did not change significantly. Wound healing and histopathological analysis showed that P.petiolosa ointment could promote wound healing in normal wound group and S.aureus infection group. ELISA results showed that P.petiolosa ointment group can significantly reduce the inflammatory response in the early stage of trauma. The results demonstrated that this selfdeveloped Chinese medicinal ointment with P.petiolosa as the main raw material has a good effect of antagonizing S.aureus and promoting wound healing.