With the continuous improvement of residents′ quality of life and aesthetic level, the ecological and visual beauty of landscape has been widely valued by all sectors of society. The research on landscape visual evaluation has gradually become the central issue of landscape aesthetics research. The landscape visual evaluation reflects the style and features of the main body of the landscape evaluation and gives people a visual feeling. Based on CNKI and Web of Science (WoS) core collection databases, this study obtained Chinese and foreign literature from 1998 to 2022, and analyzed the hot topics and research trends in the field of landscape visual evaluation using bibliometrics and literature review. The knowledge map was drawn through CiteSpace software for visual analysis and information statistics of network maps such as keywords and author organizations. The results show that: 1) The total number of foreign documents in landscape visual evaluation research was more than that of Chinese documents. From 1998 to 2008, the number of Chinese documents issued by the WoS core collection was in the vacancy stage. From 2009, the number of documents issued by the WoS core collection began to rise rapidly, and from 1998 to 2022, the number of Chinese documents issued by CNKI showed a trend of slow growth to rapid growth. 2) The research focus of Chinese literature landscape visual evaluation focuses on the landscape evaluation and visual analysis of villages, urban parks and scenic spots, while foreign literature research focuses on the landscape ecological evaluation and visual impact analysis of large-scale sites. 3) The emergence of visual evaluation systems, models and schools in Europe and the United States has provided a theoretical basis for landscape visual evaluation. In the process of research and development of landscape visual evaluation, it has shown a comprehensive integration of multiple disciplines. Modern technologies such as 3S, 3D visualization and digitization have provided new ideas and means for landscape visual evaluation. 4) The research field of landscape visual evaluation at home and abroad has formed a cooperative network of authors and institutions of a certain scale, and the cooperation between international authors and institutions is relatively closer. By summarizing and sorting out the existing research achievements in the field of landscape visual evaluation, scientific research basis is provided for establishing a scientific landscape visual evaluation system, and the operability of the evaluation work is improved to promote the effective transformation of landscape aesthetic value into engineering practice.

Multimodal sentiment analysis is an important task in the field of natural language processing, and modality fusion is its core problem. Previous research has not distinguished the primary and secondary status of each modality in sentiment analysis, treating each modality equally and not properly recognizing the quality and performance gaps between different modalities. Existing research shows that textual modalities tend to dominate sentiment analysis, but non-textual modalities contain key feature information that is essential for identifying correct sentiment. Therefore, this paper proposes a modality fusion strategy that focuses on text modality. Through a codec network with an attention mechanism to distinguish the shared and private semantics between different modalities, the two semantic enhancements of non-text modalities relative to text modalities are used to complement text features, achieve a joint robust representation of multiple modalities, and ultimately achieve sentiment prediction. Experiments on the CMU-MOSI and CMU-MOSEI video sentiment analysis datasets show that the accuracy of this method reaches 87.3% and 86.2% respectively, outperforming many existing state-of-the-art methods.

A convolutional neural network model based on multi-level attention mechanism is proposed to solve the problem that the existing methods of fake review detection do not make full use of the text features of fake reviews. Firstly, a variety of pre-trained word vectors are used to initialize the word embedding layer, and complex position coding is carried out. Then, multiple feature maps are obtained by convolution of multiple convolution kernels through the channel level and convolution kernel level attention layer embedded with user features, and different weights are assigned according to the importance of features. Finally, the feature representation of the fitted reviews text is classified by softmax. Experimental results show that compared with many mainstream excellent neural network models, the accuracy rate of the proposed model increases by 4.74%, and the F₁ value gains by 3.86%.

In order to improve the performance and robustness of point cloud classification, a point cloud classification network combining residual dynamic graph convolution and feature enhancement is proposed. Multi-direction coding method is adopted to select the nearest neighbor points in the spatial multi-direction of the local neighborhood center point to enrich the point cloud features. The residual dynamic graph convolution is used to extract features, and the residual structure is used to deeply fuse local features and global features to effectively alleviate the network degradation problem. The reinforcement spatial attention module is constructed so that the network can learn to adaptively assign weights to different neighborhood features in the spatial domain, enhance useful features, and suppress redundant features. High-level links are used to retain more feature information. Experiments show that the overall classification accuracy of the above models on the ModelNet10 and ModelNe40 data sets is 94.81% and 93.62% respectively, with higher classification accuracy and stronger robustness, which is superior to the existing advanced methods.

QAR(quick access recorder) data retains a large number of flight parameters of the aircraft, which makes it possible to study the flight safety of the aircraft and ensure the flight quality. Aiming at the requirement of multi-dimensional QAR data security detection, an effective and interpretable multi-dimensional time series anomaly detection model based on convolution VAE and multi-head self-attention-LSTM is proposed. The new model is able to capture both spatial and temporal information between multiple time series data channels, and perform interpretable reconstruction of sequential patterns to detect all types of anomalies. In the experiment, compared with the existing methods such as iForest, LSTM, VAE, LSTM-VAE, and USAD on real QAR datasets, the results show that the F1-scores of this model in three different stages of anomaly detection reach 0.891 2, 0.942 4, and 0.953 7, respectively, which are superior to other models and can accurately detect anomalies in multidimensional QAR data.

Most chemostat models ignore the wall attachment of microorganisms and little seems to be known about the memory effect on stochastic biological systems. This paper is devoted to study the existence of random attractors for chemostat models with wall attachment driven by fractional Brownian motion. Firstly, a suitable stopping time sequence is established and the continuous random dynamical system is divided into discrete random dynamical systems on the small stopping time interval. Then the random set is constructed over a closed ball and its compactness, temperedness and attractivity are proved. Moreover, the required results on the existence of random attractor for generated random dynamical system are guaranteed. Finally, numerical simulations are carried out to verify the correctness and effectiveness of the theoretical results obtained.

The existence of radial symmetric solutions of a class of p-Laplace equation on the exterior domain Ω={x∈R^N:x>r₀}(N≥2) is investigated in this paper. For continuous function f:R→R, unlike the previous literatures, the condition that f is nonegative is removed. Under the condition that f satisfies an appropriate inequality, an existence result of radial symmetric solutions is obtained by applying the Leray-Schauder fixed point theorem. On this basis, the uniqueness of radial symmetric solutions is discussed.

Based on the PI controller design theory, this paper firstly analyzes the stability of power system load frequency and proposes an improved event triggering mechanism, then, processes the mathematical mean expectation of power system with uncertain semi-Markov jump parameters, and finally constructs a closed-loop system that can adjust adaptively with the change of sampling signal state. Considering the interference effect and time-delay virtual delay in the system with random jump multimode, a modal-dependent controller is designed. Firstly, the functional derivative integral term is scaled by combining matrix cross product inequality and Schur’s complement theorem, so that the optimal value problem is transformed into solving linear matrix inequality. Then, the control gain and event trigger matrix are solved by matlab toolbox, and the stochastic stability criterion of the system is given. Finally, the performance index of the system H_∞ is analyzed. The designed scheme can improve the transmission rate of control signals in the communication network, effectively reduce the burden of power grid information transmission, save energy and improve the dynamic performance of the system.

As ultra-low dielectric layer dielectric (ILD) materials, metal-organic frameworks (MOFs) show great application potential in the development of large-scale integrated circuit for high-end chips and other microelectronic devices. In this paper, MOF-2 with good crystallinity was prepared in the aqueous phase at room temperature and pressure, and the lattice parameters were analyzed by Rietveld refinement. Because of molecular structure rearrangement, the crystal structure of MOF-2 was changed into two desolvent products with [Zn(BDC)(DMF)] and [Zn(BDC)(H₂O)] during thermal dehydration. After heating at 100 ℃ for 6 h, the phases of these two products were 89% and 11% respectively. The dielectric properties of the products were studied by using dielectric spectrum. It was found that the dielectric constant of MOF-2 prepared in aqueous phase changed greatly with frequency, and the relative dielectric constant was 4-5 above 1 MHz. The synergistic effect of water and DMF molecule had a major impact on the dielectric response of MOF-2 at mid and low frequency. The relative dielectric constant of the product no longer changes with the frequency after heating and remains at about 3.39 (0.1-1 MHz). After removing all solvent molecules, MOF-2 generates [Zn(BDC)] with sheet structure, and the relative dielectric constant is 3.91 (0.1-1 MHz). Based on the dielectric relaxation theory, the dielectric response of MOF-2 could be described by four relaxation mechanisms: dc conductivity in low frequency, synergistic polarization of water molecules and DMF in intermediate frequency, electron/ion displacement polarization in high frequency, and the universal dielectric response (UDR). This study clarified the structural characteristics, dielectric properties and polarization relaxation mechanism of MOF-2 synthesized in aqueous phase and its desolvent products, the result provided a theoretical basis for the application of MOF-2 in the electrical field.

The chemical constituents of Gymnema sylvestre (Retz.) Schult. were isolated and purified by macroporous adsorption resin, silica gel, Sephadex LH-20 and semipreparative high performance liquid chromatography. Their structures were identified by physicochemical properties and spectral data. Sixteen compounds were isolated and identified as sylvepregoside E (1), gymnepregoside H (2), apigenin (3), luteolin (4), phydroxybenzaldehyde (5), phydroxybenzoic acid (6), cinnamic acid (7), caffeic acid (8), 3, 4-dimethoxycinnamic acid (9), transphydroxycinnamic acid ethyl ester (10), 4-O-β-D-glucopyranosylvanillic acid (11), ochuscin C (12), pinoresinol (13), 4-hydroxy-2-nonenoic acid (14), linolenic acid (15), and linoleic acid (16). Among them, compound 1 was a new compound, compounds 2-16 were isolated from this plant for the first time. The results of biological activity showed that compounds 1 and 2 could significantly promote the glucose uptake of L6 cells, which were 1.79 and 1.95 times higher than that of the blank group, respectively.

Red mud-sodium alginate hydrogel (RMSA) was prepared by cross-linking reaction using red mud and sodium alginate as raw materials. The effects of initial pH of solution, adsorption temperature, adsorption time, initial concentration and multiple heavy metal systems on the adsorption effect of RMSA on Pb(Ⅱ) were investigated by batch experiments and combined with XRD, FTIR and SEM-EDS characterization analysis, its adsorption characteristics on Pb(Ⅱ) was studied. The experimental results showed that the proposed secondary kinetic adsorption model and Langmuir adsorption isotherm model could better describe the adsorption process of RMSA on Pb(Ⅱ), which belonged to single molecular layer chemisorption. The maximum theoretical adsorption capacity of RMSA on Pb(Ⅱ) was 454.54 mg/g under the optimal conditions of pH=6, temperature of 25 ℃, adsorption time of 900 min and initial concentration of Pb(Ⅱ) of 30-900 mg/L. The adsorption of RMSA on Pb(Ⅱ) was more selective in the adsorption experiments of multivariate heavy metal systems. The analysis showed that ion exchange was the main mechanism of Pb(Ⅱ) adsorption by RMSA. In addition, RMSA still maintained high adsorption performance in the five-cycle experiment, which had a good application prospect in terms of economic applicability.

From January to December, in the year 2015, camera trapping method was used to assess the distribution of Taihangshan Macaque (Macaca mulatta tcheliensis) and the seasonality of activity rhythms in the Taihangshan Macaque National Nature Reserve in Henan Jiyuan. The results showed that Taihangshan Macaques were distributed at 37 sites in three forest farms. The daily activity of the macaques had two peaks throughout the year with the morning peak at 6:40-9:30 and the afternoon peak at 15:20-18:20. Compared with that in other seasons, the morning peak in the summer was 0.5-3.2 h earlier, and its peak period was extended by 0.3-2.4 h. In addition, human disturbance was distributed at 66 sites in four forest farms. The daily peak of human activities occurred at 10:00-18:00 throughout the year. Compared with that in other seasons, the peak activity in the summer was 0.5-2.8 h earlier, and the peak activity was extended by 0.5-1.7 h. In sum, the results indicated that the seasonal variation of daily activity rhythms from this macaque are mainly affected by factors such as temperature, sunshine intensity and sunrise time, and are influenced by human activities to a certain extent.

Protecting the habitats of endangered wildlife is the basis for intensive research, development and protection of biological resources, which is one of the key tasks in promoting the construction of ecological civilization. The endemic birds, Tragopan caboti has been included in the list of vulnerable species in the International Union for Conservation of Nature for habitat loss and fragmentation. Based on 298 distribution screened records and 12 environmental factors, the potentially suitable habitats of the T. caboti was predicted using MaxEnt model. The results showed that: 1) The distribution of the T. caboti was mainly affected by precipitation, atmospheric temperature and vegetation types. 2) The total potentially habitats area of the T. caboti was 66.76×10⁴ km². Specifically, the low suitable habitats, medium suitable habitats and high suitable habitats were about 33.38×10⁴ km², 20.04×10⁴ km², 13.34×10⁴ km², respectively. 3) The distribution of potential habitats had a high overlap with the actual distribution density. Furthermore, the high suitable habitats were mainly concentrated in Northern Fujian and Northeast Guangxi. Therefore, great attention should be paid to the conservation of T. caboti’s highly-suitable habitat, such as in northern Fujian and northeast Guangxi, China.

In order to reveal the distribution pattern of ant species in the northern part of the western Sichuan Plateau, the distribution pattern of ant species in the northern part of this region was studied by plot survey. The results showed that 35 ant species, belonging to 17 genera and 4 subfamilies, lived in the northern part of the western Sichuan Plateau. Formica sentschuensis Ruzsky, F. fusca Linnaeus, Myrmica kozlovi Ruzsky and M. jessensis Forel were the most widely distributed and adaptable species in the northern part of the Western Sichuan Plateau. Ants showed obvious differentiation in habitat, foraging and nesting sites; and the broadleaved forest had the most abundant species; among the foraging sites, surface foraging was the most abundant; nesting sites were mainly under the stone and in the soil. In conclusion, the distribution pattern of ant species was mainly related to the adapt ability of species, but also affected by altitude, vegetation and anthropogenic disturbance. The results are of great significance to enrich and improve the diversity of ants in the southwest mountains of China.

Pheromones can be recognized by Procambarus clarkii in water for courtship. Clarifying the effective components of P. clarkii pheromones can provide a theoretical basis for green management of aquatic invasive P. clarkii. In this research, the effects of eight standard compounds, namely 4-thiouridine, bilirubin, ethinyl estradiol, 17α-hydroxyprogesterone caproate, N-acetyl-L-leucine, GTP disodium salt, and GDP disodium salt and GMP disodium salt, on the selection behavior of P. clarkii were tested using a Y-shaped water maze, and then screened for the eliciting effect of ethinyl estradiol and 17α-hydroxyprogesterone caproate to prepare concentration assays. The results showed that the male P. clarkii showed nearly 50% selective response rate to 4-thiouridine, bilirubin, N-acetyl-L-leucine, GTP disodium salt, and GDP disodium salt and GMP disodium salt, with selection coefficients between ±0.15 and insignificant luring effects (P>0.05 ). P. clarkii showed high selection response rates to ethinyl estradiol (64.10%) and 17α-hydroxyprogesterone caproate (63.24%) with selection coefficients of 0.28 and 0.26, which showed significant sex differences. The minimum effective luring threshold of ethinyl estradiol and 17α-hydroxyprogesterone caproate to male P. clarkii were 10 and 100 μmol/L, respectively, while female P. clarkii were not sensitive to both compounds. Thus, ethinyl estradiol and 17α-hydroxyprogesterone caproate had significant luring effects on adult male P. clarkii but no significant response to female crayfish, which tentatively showed that ethinyl estradiol and 17α-hydroxyprogesterone caproate were the active components of the sex pheromone in P. clarkii.

Molecular marker-assisted selection breeding is an important method to obtain genes controlling target traits. In this study, four SSR markers (PCZ002, PCZ023, PCZ142 and PCZ187) associated with the growth traits of P. massoniana were selected from the second-generation full-sib families of P. massoniana. Four hybrid families were selected for each marker, with 150, 172, 157 and 163 individual materials, respectively. The association analysis of the second-generation full-sib population of P. massoniana was carried out by combining the data of tree height growth traits. The results showed that the genotypes of PCZ142-AA and PCZ142-AB were positively correlated with tree height, and the tree height was significantly higher than that of PCZ142-BB genotype, while PCZ142-BB genotype was significantly negatively correlated with tree height (P< 0.05). There was a significant positive correlation between PCZ187-BB genotype and tree height (P< 0.01), and the tree height was significantly higher than that of PCZ187-AA and PCZ187-AB genotype materials. The genotypes marked PCZ002 and PCZ023 had no significant correlation with tree height. In summary, the markers PCZ142-BB and PCZ187-BB were significantly correlated with the height growth traits of P. massoniana. The PCZ142-BB was significantly negatively correlated, and the PCZ187-BB was significantly positively correlated. These two genotype materials could become important genetic resources for the height growth traits of P. massoniana. The results of this study provide important information for the locus of height growth traits of P. massoniana in the positioning control population, and provide reference for molecular marker-assisted selection breeding of P. massoniana.

Non-targeted mass spectrometry analysis based on ultra high performance liquid chromatography-quadrupole-time of flight-tandem mass spectrometry (UPLC-Q-TOF-MS) combined with machine learning can predict whether beer is infected by spoilage bacteria. The metabonomics was used to analyze beer spoilage bacteria such as Lactobacillus brevis, Bacillus, Lactobacillus parabuchneri, Lactobacillus plantarum and Staphylococcus saprophyticus during beer fermentation. The metabonomics data was then analyzed by the multivariate analysis method of random forest. The first 30% of the important characteristic substances of the model were selected for pathway enrichment analysis. Finally, it was found that in the initial stage of fermentation, the contents of choline, choline glycerophosphate, triethanolamine, phosphatidylethanolamine and 2-(3-carboxylpropionyl)-6-hydroxy-cyclohexa-2,4-dicarboxylic acid (SHCHC) in spoiled beer samples were higher than those in normal beer samples. The contents of choline in spoiled beer with 36-48 h fermentation were 50% - 130% higher than those in normal beer samples; The content of glycerophosphate choline in spoiled beer with 24 h fermentation was 10% - 30% lower than that in normal beer. As the fermentation goes on, the content of glycerophosphate choline in spoiled beer gradually increased to more than 9% higher than that in normal beer; The content of triethanolamine in normal beer with 24 h fermentation was the highest, which was 2-3 times higher than that in spoiled beer; Phosphatidylethanolamine in spoiled beer was 6% - 13% higher than that in normal beer with 96 h fermentation; SHCHC was also only detected in spoiled beer with 36 h fermentation. These substances could be used as biomarkers of beer spoilage.

By studying four typical land use types in the east of Hainan Island, the distribution characteristics of soil organic carbon content and the change of carbon density under different land use patterns were analyzed. In order to provide a scientific basis for reasonably optimizing the land use mode of Hainan Island and promoting the sustainable development of agricultural production. The study took the east of Hainan Island as the research area, selected four typical land use types (paddy field, abandoned land, orchard land and rubber forest land) as the research objects. Through field investigation and sampling and indoor test analysis, the distribution characteristics of soil organic carbon content and density under different land use types were studied. The change of land use types can lead to the change of soil bulk density. The results showed that: the average soil organic carbon content of 0-30 cm under four different land use types in the study area ranged from 3.73 to 14.01 g/kg. The average soil organic carbon content of paddy field was the highest, and that of rubber plantation was the lowest. The average soil organic carbon content from large to small was paddy field > abandoned land >orchard > rubber plantation. The average organic carbon density of 0-30 cm soil ranged from 1.74 to 5.86 kg·m^-2; Under the same land use types, the change of soil organic carbon density in each soil layer was basically the same. The change trend of soil organic carbon density and soil organic carbon content of four different land use types were similar, and they all decreased with the increase of soil depth. The soil organic carbon density of the four land use types in 0-20 cm soil layer accounted for more than 65% of the study section (0-30 cm) (69.21%-72.99%). The change of land use types would lead to the change of soil bulk density .Among the four different land use types in the study area, the content and density of soil organic carbon in paddy field were the highest, and the content and density of rubber forest land were the lowest; There were differences in soil carbon input under different land use types, mainly because land use types could affect the exogenous input of soil carbon pool; Under the same natural conditions, there were significant differences in soil organic carbon content due to different land use types. Therefore, Hainan should strengthen the management of rubber forest land, especially the use of organic fertilizer.

Karst wetland plays an important ecological role in karst area, which is widely distributed in karst area. It is of great significance to explore the influencing factors of habitat quality change in karst wetland for ecological construction in karst area. The data used in this study were the land use data of Huixian Wetland Reserve in 2000 and 2020. The InVEST model was used to assess the habitat quality of the Huixian Wetland Reserve, and the geographic detector was employed to detect the influencing factors of habitat quality change in the Huixian Wetland Reserve. The results showed that:1) The construction land in the study area expanded significantly from 2000 to 2020, while the grassland and water area decreased significantly. 2)The habitat quality of the Huixian Wetland Reserve showed an overall decreasing trend, and the degree of habitat degradation had increased, with the areas of poor habitat quality mainly concentrated in the northern, central and northeastern parts of the study area. 3)The results of factor detection and interaction detection showed that changes in habitat quality in the Huixian Wetland Reserve were influenced by both natural and anthropogenic factors, with the anthropogenic factor being the dominant influence factor GDP (q of 0.135), followed by the nighttime light index (DMSP) (q of 0.053). The interaction detection results showed that the interaction of any two factors was greater than the sum of the effects of two single factors, showing a two-factor enhancement effect. The results are helpful to reveal the spatio-temporal evolution process of habitat quality in Huixian Wetland Reserve, and to put forward feasible strategies for ecological construction in Huixian Wetland Reserve.