Swing Trend Prediction of Main Guide Bearing in Hydropower Units Based on MFS-DCGNN.

Hydropower units are the core equipment of hydropower stations, and research on the fault prediction and health management of these units can help improve their safety, stability, and the level of reliable operation and can effectively reduce costs. Therefore, it is necessary to predict the swing trend of these units. Firstly, this study considers the influence of various factors, such as electrical, mechanical, and hydraulic swing factors, on the swing signal of the main guide bearing y-axis. Before swing trend prediction, the multi-index feature selection algorithm is used to obtain suitable state variables, and the low-dimensional effective feature subset is obtained using the Pearson correlation coefficient and distance correlation coefficient algorithms. Secondly, the dilated convolution graph neural network (DCGNN) algorithm, with a dilated convolution graph, is used to predict the swing trend of the main guide bearing. Existing GNN methods rely heavily on predefined graph structures for prediction. The DCGNN algorithm can solve the problem of spatial dependence between variables without defining the graph structure and provides the adjacency matrix of the graph learning layer simulation, avoiding the over-smoothing problem often seen in graph convolutional networks; furthermore, it effectively improves the prediction accuracy. The experimental results showed that, compared with the RNN-GRU, LSTNet, and TAP-LSTM algorithms, the MAEs of the DCGNN algorithm decreased by 6.05%, 6.32%, and 3.04%; the RMSEs decreased by 9.21%, 9.01%, and 2.83%; and the CORR values increased by 0.63%, 1.05%, and 0.37%, respectively. Thus, the prediction accuracy was effectively improved.

Investigating the environmental capacity of soil heavy metals and its determinants in agro-pastoral regions of the qinghai-tibetan plateau.

Environmental capacity (EC) serves as the basis for environmental planning and management, as a key indicator for assessing environmental risk and quality, and as a foundation for achieving sustainable development. Studies on EC typically address agricultural or urban rather than pastoral areas, with few examining agro-pastoral areas. The EC of the Tibetan Plateau is particularly important, considering its importance as an agricultural area and ecological reserve. To address this gap, the Qingshizui area in Menyuan County, a typical agro-pastoral area on the Tibetan Plateau, was selected to quantify soil EC and its spatial distribution. In terms of the dynamic and static annual soil EC for this region, the heavy metals were ranked as follows, in ascending order: Cd, Hg, Co, As, Sb, Ni, Cu, Pb, Cr, and Zn. Most of the areas with high residual EC were in the west. For the 10 heavy metals, residual EC was significantly affected by geological background. For all the heavy metals except Zn and Hg, residual EC was significantly affected by soil type. The heavy metal elements in the agro-pastoral area's soil are mildly enriched, suggesting minimal human impact. The composite EC index of this soil is 0.98, indicating an intermediate EC and low health risk. This study underscores that integrating agriculture and pastoralism can optimize land use and mitigate ecological pressures associated with these practices when done separately. Our research provides valuable insights for resource optimization, environmental conservation, and enhancing the welfare of farmers and herders in the Qinghai-Tibet region.

A Deep Learning Model with Signal Decomposition and Informer Network for Equipment Vibration Trend Prediction.

Accurate equipment operation trend prediction plays an important role in ensuring the safe operation of equipment and reducing maintenance costs. Therefore, monitoring the equipment vibration and predicting the time series of the vibration trend is one of the effective means to prevent equipment failures. In order to reduce the error of equipment operation trend prediction, this paper proposes a method for equipment operation trend prediction based on a combination of signal decomposition and an Informer prediction model. Aiming at the problem of high noise in vibration signals, which makes it difficult to obtain intrinsic characteristics when directly using raw data for prediction, the original signal is decomposed once using the variational mode decomposition (VMD) algorithm optimized by the improved sparrow search algorithm (ISSA) to obtain the intrinsic mode function (IMF) for different frequencies and calculate the fuzzy entropy. The improved adaptive white noise complete set empirical mode decomposition (ICEEMDAN) is used to decompose the components with the largest fuzzy entropy to obtain a series of intrinsic mode components, fully combining the advantages of the Informer model in processing long time series, and predict equipment operation trend data. Input all subsequences into the Informer model and reconstruct the results to obtain the predicted results. The experimental results indicate that the proposed method can effectively improve the accuracy of equipment operation trend prediction compared to other models.

A Multistep Prediction Model for the Vibration Trends of Hydroelectric Generator Units Based on Variational Mode Decomposition and Stochastic Configuration Networks.

Accurately predicting the changes in turbine vibration trends is a key part of the operational condition maintenance of hydropower units, which is of great significance for improving both the operational condition and operational efficiency of hydropower plants. In this paper, we propose a multistep prediction model for the vibration trend of a hydropower unit. This model is based on the theoretical principles of signal processing and machine learning, incorporating variational mode decomposition (VMD), stochastic configuration networks (SCNs), and the recursive strategy. Firstly, in view of the severe fluctuations of the vibration signal of the unit, this paper decomposes the unit vibration data into intrinsic mode function (IMF) components of different frequencies by VMD, which effectively alleviates the instability of the vibration trend. Secondly, an SCN model is used to predict different IMF components. Then, the predicted values of all the IMF components are superimposed to form the prediction results. Finally, according to the recursive strategy, a multistep prediction model of the HGU's vibration trends is constructed by adding new input variables to the prediction results. This model is applied to the prediction of vibration data from different components of a unit, and the experimental results show that the proposed multistep prediction model can accurately predict the vibration trend of the unit. The proposed multistep prediction model of the vibration trends of hydropower units is of great significance in guiding power plants to adjust their control strategies to reach optimal operating efficiency.

A Dynamic Monitoring Method of Public Opinion Risk of Overseas Direct Investment-Based on Multifractal Situation Optimization.

The negative public opinions and views on overseas direct investment (ODI) of a multinational enterprise (MNE) will damage the image of its brand and are likely to bring it serious economic and social losses. So, it is important for the MNE to understand the formation and spread mechanism of public opinion risk (POR) in order to effectively respond to and guide the public opinion. This research proposed a multifractal-based situation optimization method to explore the POR evolution based on the media-based negative sentiment on China's ODI. The sentiment measurement is obtained by a directed crawler for gathering the text of media reports corresponding to a certain ODI event using a URL knowledge base from the GDELT Event Database. Taking the public opinion crisis of the tax evasion incident of the local arm of China's MNE in India as an example, the experiments show that this method could dynamically monitor the POR event in real-time and help MNE guide the effective control and benign evolution of public opinion of the event.

Spatio-temporal evolution and trend prediction of the incidence of Class B notifiable infectious diseases in China: a sample of statistical data from 2007 to 2020.

BACKGROUND: With the accelerated global integration and the impact of climatic, ecological and social environmental changes, China will continue to face the challenge of the outbreak and spread of emerging infectious diseases and traditional ones. This study aims to explore the spatial and temporal evolutionary characteristics of the incidence of Class B notifiable infectious diseases in China from 2007 to 2020, and to forecast the trend of it as well. Hopefully, it will provide a reference for the formulation of infectious disease prevention and control strategies. METHODS: Data on the incidence rates of Class B notifiable infectious diseases in 31 provinces, municipalities and autonomous regions of China from 2007 to 2020 were collected for the prediction of the spatio-temporal evolution and spatial correlation as well as the incidence of Class B notifiable infectious diseases in China based on global spatial autocorrelation and Autoregressive Integrated Moving Average (ARIMA). RESULTS: From 2007 to 2020, the national incidence rate of Class B notifiable infectious diseases (from 272.37 per 100,000 in 2007 to 190.35 per 100,000 in 2020) decreases year by year, and the spatial distribution shows an "east-central-west" stepwise increase. From 2007 to 2020, the spatial clustering of the incidence of Class B notifiable infectious diseases is significant and increasing year by year (Moran's I index values range from 0.189 to 0.332, p < 0.05). The forecasted incidence rates of Class B notifiable infectious diseases nationwide from 2021 to 2024 (205.26/100,000, 199.95/100,000, 194.74/100,000 and 189.62/100,000) as well as the forecasted values for most regions show a downward trend, with only some regions (Guangdong, Hunan, Hainan, Tibet, Guangxi and Guizhou) showing an increasing trend year by year. CONCLUSIONS: The current study found that since there were significant regional disparities in the prevention and control of infectious diseases in China between 2007 and 2020, the reduction of the incidence of Class B notifiable infectious diseases requires the joint efforts of the surrounding provinces. Besides, special attention should be paid to provinces with an increasing trend in the incidence of Class B notifiable infectious diseases to prevent the re-emergence of certain traditional infectious diseases in a particular province or even the whole country, as well as the outbreak and spread of emerging infectious diseases.

An Ensemble Prognostic Method of Francis Turbine Units Using Low-Quality Data under Variable Operating Conditions.

The prognostic is the key to the state-based maintenance of Francis turbine units (FTUs), which consists of performance state evaluation and degradation trend prediction. In practical engineering environments, there are three significant difficulties: low data quality, complex variable operation conditions, and prediction model parameter optimization. In order to effectively solve the above three problems, an ensemble prognostic method of FTUs using low-quality data under variable operation conditions is proposed in this study. Firstly, to consider the operation condition parameters, the running data set of the FTU is constructed by the water head, active power, and vibration amplitude of the top cover. Then, to improve the robustness of the proposed model against anomaly data, the density-based spatial clustering of applications with noise (DBSCAN) is introduced to clean outliers and singularities in the raw running data set. Next, considering the randomness of the monitoring data, the healthy state model based on the Gaussian mixture model is constructed, and the negative log-likelihood probability is calculated as the performance degradation indicator (PDI). Furthermore, to predict the trend of PDIs with confidence interval and automatically optimize the prediction model on both accuracy and certainty, the multiobjective prediction model is proposed based on the non-dominated sorting genetic algorithm and Gaussian process regression. Finally, monitoring data from an actual large FTU was used for effectiveness verification. The stability and smoothness of the PDI curve are improved by 3.2 times and 1.9 times, respectively, by DBSCAN compared with 3-sigma. The root-mean-squared error, the prediction interval normalized average, the prediction interval coverage probability, the mean absolute percentage error, and the R2 score of the proposed method achieved 0.223, 0.289, 1.000, 0.641%, and 0.974, respectively. The comparison experiments demonstrate that the proposed method is more robust to low-quality data and has better accuracy, certainty, and reliability for the prognostic of the FTU under complex operating conditions.

Database comments on Telegram channels related to cryptocurrencies with sentiments.

OBJECTIVES: Due to the limitations of Twitter, the expansion of Telegram channels, and the Telegram API's easy use, Telegram comments have become prevalent. Telegram is one of the most popular social networks, unlike Twitter, which has no restrictions on sending messages, and experts can share their opinions and media. Some of these channels, managed by influencers of large companies, are very influential in the behavior of the market on various stocks, including cryptocurrencies. In this research, the opinion collection of 10 famous Telegram channels regarding the analysis of cryptocurrencies has been extracted. The sentiments of these opinions have been analyzed using the HDRB model. HDRB is a hybrid model of RoBERTa deep neural network, BiGRU, and attention layer used for sentiment analysis (SA). Analyzing the sentiments of these opinions is very important for understanding the future behavior of the market and managing the stock portfolio. The opinions of this dataset, published by experts in the field of cryptocurrencies, are precious, unlike the opinions that are extracted only by using the hashtag of the names of cryptocurrencies. On the other hand, the dataset related to cryptocurrencies, which has the opinions of experts and the polarity of their feelings, is very rare. DATA DESCRIPTION: The dataset of this research is the sentiments of more than ten popular Telegram channels regarding a wide range of cryptocurrencies. These comments were collected through the Telegram API from December 2023 to March 2024. This data set contains an Excel file containing the text of the comments, the date of comment creation, the number of views, the compound score, the sentiment score, and the type of sentiment polarity. These opinions cover influencer analysis on a wide range of cryptocurrencies. Also, two Word files, one containing the description of the dataset columns and the other Python code for extracting comments from Telegram channels, are included in this dataset.

Database of twitter influencers in cryptocurrency (2021-2023) with sentiments.

OBJECTIVES: With the expansion of social networks such as Twitter, many experts share their opinions on various topics. The opinions of experts, who are also known as influencers, can be very influential. Combining these tweets and the historical prices of cryptocurrencies makes it possible to predict their price trends accurately. A Hybrid of RoBERTa deep neural network and BiGRU has been used for Sentiment Analysis (SA). Sentiments of tweets can be of great help to investors to understand the future behavior of the market and manage the stock portfolio. Unlike the tweets that are only extracted using the cryptocurrency name hashtag, the tweets of this dataset have specialized opinions and can determine the market trend. DATA DESCRIPTION: The dataset created in this research concerns the opinions of more than 52 influencers (persons or companies) regarding eight cryptocurrencies. This dataset was collected through the Apify Twitter API for eight months, from February 2021 to June 2023. This dataset contains five Excel files and tweets, compound score, importance coefficient of each tweet, sentiment polarity, and historical prices of four cryptocurrencies: Bitcoin, Ethereum, Binance, and other information. These tweets cover the opinions of 52 influencers on more than 300 cryptocurrencies, although most comments are related to Bitcoin, Ethereum, and Binance. For this reason, three Excel files containing the historical prices of polarity and compound sentiment related to Bitcoin, Ethereum, and Binance cryptocurrencies have been placed separately in the dataset. The polarity of sentiment in these Excel shows the maximum number of polarities by applying the importance coefficient, which determines the dominant polarity of sentiment related to a particular day for the cryptocurrency.

Regional differences, dynamic evolution and trend prediction of green manufacturing development levels in China.

Green manufacturing has become a necessary way to promote new industrialization and realize the high-quality development of China's manufacturing industry. Based on the panel data of 30 provinces in China from 2012 to 2022, this paper constructs a comprehensive evaluation index system for the green manufacturing development level and introduces the TOPSIS- Gray correlation method to comprehensively measure the green manufacturing development level of China as a whole and the four major regions in the eastern, central, western, and northeastern parts of the country. The regional differences, distribution dynamics and evolutionary trends of China's green manufacturing development level are also explored with the help of the Gini coefficient, kernel density estimation and Markov chain methods. Research Findings: (1) The green manufacturing development level in China is on an upward trend, with an overall spatial distribution pattern of "East is superior and West is inferior". (2) There are regional differences in the green manufacturing development level in China, and the differences are widening, with interregional differences being the main reason for this overall difference. (3) The country as a whole, the central region and the western region are polarized to varying degrees, with the rest of the country showing an improvement in polarization. (4) Without considering spatial factors, the development of green manufacturing in each province experiences "club convergence" in the short term, and it is difficult to realize rapid development. Considering spatial factors, China's green manufacturing development level is generally characterized by "elevated in proximity to high levels and suppressed in proximity to low levels", and in the long run, it shows a distribution trend toward the concentration of high values. The findings of this study can provide new ideas for promoting synergistic efficient development of green manufacturing in China.

Mechanism-based cytotoxicity trend prediction of furan-containing pollutants present in a mixture.

Human exposure to furan-containing pollutants (FCPs) has raised concerns due to their high risk of toxicity. A substantial number of approximately 8500 recorded compounds containing a furan ring exist which have been analytically or in biologically studied. A significant portion of these compounds is found in the everyday environments of individuals, particularly when ingested through food. Consequently, there is a need for a universal approach to rapidly predict the potential toxicity trends of FCPs. In this study, we developed a bromine labeling-based platform that combines LC-ICP-MS and LC-ESI-MS techniques to absolutely quantify FCP-induced protein adduction. The LC-ESI-MS approach facilitated the identification of FCP-derived protein adducts and optimized liquid chromatographic conditions for analyte separation. By employing a well-designed bromine-containing compound as a general internal standard, LC-ICP-MS-based technique enabled to absolutely assess bromine-labeled protein adduction. The protein adduction efficiencies of furan, 2-methylfuran, and 2,5-dimethylfuran were found to be 2.68, 2.90, and 0.37 molecules per 10,000 FCP molecules that primary hepatocytes received, respectively. Furthermore, we observed that 2-methylfuran exhibited the highest cytotoxicity, followed by furan and 2,5-dimethylfuran, which aligned with the order of their protein adduction. Thus, the protein adduction efficiency of FCPs could serve as a potential index for predicting their toxicity trends.

Physical mechanism-corrected degradation trend prediction network under data missing.

Accurate degradation trend prediction (DTP) is crucial for optimizing equipment operation and maintenance, thereby boosting production efficiency. This study introduces a novel Data Repair and Dual-data-stream LSTM (DR-DLSTM) network to tackle the challenge of missing data in equipment DTP. The proposed DR-DLSTM framework employs convex optimization to consider both the trend and periodic variations in the data, incorporating polynomial and trigonometric functions into the implicit feature matrix to construct latent vectors for missing data rectification. The network features a Dual-LSTM block with dual data streams to enhance feature extraction, with two gating update units correlating time series components and redistributing feature weights. The Dual-LSTM enables separate and accurate prediction of trend and periodic components, thereby enhancing the feature extraction capability of the prediction model. Additionally, the integration of physical rule information through Fourier and wavelet transform frequency correction modules allows for dynamic adjustments in prediction outcomes, from global trends to localized details. The DR-DLSTM's effectiveness is demonstrated through comprehensive comparisons with state-of-the-art models across multiple datasets, highlighting its superior performance. The results demonstrate the superiority of the proposed model. These algorithms were implemented in Python using Torch on a 2.9 GHz Intel I7 CPU and TITAN Xp GPU.

Intelligent alarm system for river embankment seepage based on BILSTM.

Currently, the alarm functions of existing levee seepage monitoring systems are limited to single-parameter monitoring and lack rate-of-change alarms and correlation alarms. This can lead to false alarms, missed alarms, equipment failures, or unnecessary downtime. To enhance the intelligence of levee safety monitoring and seepage alarms, a levee seepage intelligent alarm system based on a Bidirectional Long Short-Term Memory (BILSTM) network model was designed and implemented. Firstly, data cleaning and preprocessing are carried out on the engineering safety monitoring operation data to reduce the influence of dirty data such as outliers and repetitive values on the accuracy of alarms. Secondly, for the correlation between the piezometric tube water levels of the levee and the Yangtze River water levels, a correlation analysis based on Mutual Information (MI) theory was conducted to minimize the effect of piezometric tube water level change delays on correlation. Finally, the BILSTM model was used to predict trends in these potentially abnormal data intervals. Based on engineering application requirements, alarm thresholds were established, and a multi-level alarm module was developed. Field operation test results show that the proposed method can accurately predict the piezometric tube water levels of levees, achieving intelligent alarms within the engineering safety monitoring system.

Spatio-temporal evolution of social-ecological system resilience in ethnic tourism destinations in mountainous areas and trend prediction: a case study in Wuling, China.

Mountainous ethnic tourism lands are important social-ecological system types. With tourism as the main disturbance factor, the theory of social-ecological system resilience provides a new way to realize the sustainable development of ethno-tourism in mountainous areas. This study divides the social-ecological system into social, economic, and ecological subsystems. It constructs an evaluation index system to assess the resilience of ethnic tourism destinations in mountainous areas, considering vulnerability and adaptability. We investigate 64 counties in the Wuling Mountain area and use set-pair analysis to assess the resilience index of the social-ecological system from 2000 to 2020 and reveal the temporal and spatial characteristics. Obstacle degree models and a genetic algorithm-back propagation neural network are utilized to determine the influencing factors and predict future development trends. The following results were obtained: (1) Temporally, the resilience index shows a steady upward trend, reaching a moderate level. The resilience of the social subsystem fluctuates and rises; the economic subsystem exhibits slow, fast, and slow growth rates with occasional abrupt changes; and the ecological subsystem demonstrates a stable, slightly increasing trend. (2) Spatially, the resilience index is high at the edges and low in the central area, exhibiting a concave distribution. Most counties have moderate or higher resilience. The social and ecological subsystems have low resilience in the south and high resilience in the north. The resilience of the economic subsystem is high at the edges and low in the central area. (3) On the distribution of major obstacle factors, the first two are similar at the county level, and the last three are significantly different. The similarity of the barrier factors is related to the degree of regional proximity of the county, and overall, the similarity is decreasing from north to south and from west to east in the distribution pattern within the area. and to a certain extent, it is affected by terrain and geomorphology. (4) The spatial distribution of the resilience index is similar in 2025 and 2030. The index decreases slightly and then increases annually, with a lower growth rate in the south than in the north. Lower values occur in the northern and southwestern parts, whereas higher values are observed around high-value areas. The region as a whole will develop in a coordinated and integrated manner in the future.

A modeling pipeline to relate municipal wastewater surveillance and regional public health data.

As COVID-19 becomes endemic, public health departments benefit from improved passive indicators, which are independent of voluntary testing data, to estimate the prevalence of COVID-19 in local communities. Quantification of SARS-CoV-2 RNA from wastewater has the potential to be a powerful passive indicator. However, connecting measured SARS-CoV-2 RNA to community prevalence is challenging due to the high noise typical of environmental samples. We have developed a generalized pipeline using in- and out-of-sample model selection to test the ability of different correction models to reduce the variance in wastewater measurements and applied it to data collected from treatment plants in the Chicago area. We built and compared a set of multi-linear regression models, which incorporate pepper mild mottle virus (PMMoV) as a population biomarker, Bovine coronavirus (BCoV) as a recovery control, and wastewater system flow rate into a corrected estimate for SARS-CoV-2 RNA concentration. For our data, models with BCoV performed better than those with PMMoV, but the pipeline should be used to reevaluate any new data set as the sources of variance may change across locations, lab methods, and disease states. Using our best-fit model, we investigated the utility of RNA measurements in wastewater as a leading indicator of COVID-19 trends. We did this in a rolling manner for corrected wastewater data and for other prevalence indicators and statistically compared the temporal relationship between new increases in the wastewater data and those in other prevalence indicators. We found that wastewater trends often lead other COVID-19 indicators in predicting new surges.

Analysis of the structure and trend prediction of China's total health expenditure.

Background: In the context of rapid economic and social development, there has been a continuous intensification of population aging, transformation of disease patterns, and wide application of new medical technologies. As a result, health expenditures in various countries have sharply soared. How to utilize limited medical resources to maximize the improvement of health levels has become a hot and challenging issue related to the well-being of all humanity. The relevant indicators of total health expenditure play a crucial role in monitoring and evaluating the fairness of health financing and health security in the region. Objective: This study explores the changes in the main expenses that constitute China's total health expenditure and uses indicators related to health expenditure to observe the changes and future development trends of China's health expenditure. Based on this, the utilization of China's health expenditure is monitored to identify possible problems, and thereby targeted suggestions for promoting the development of China's health and wellness cause are put forward. Methods: Based on the comparison of previous literature, this paper analyzes the changes and future development trends in China's health expenditure by using the relevant indicators of China's health expenditure through the structural variation analysis method and the gray prediction model. Results: The results show that the scale of government, social, and out-of-pocket health expenditures has continuously expanded, with social health expenditures becoming the main funding source for total health expenditures. The burden of medical expenditures on individuals has been further reduced. In the institutional method of total health expenditures, hospital expenditures account for about 60% of the total and are the main component. The expenditures of health administration and medical insurance management institutions are the main driving force behind the growth of total health expenditures. However, the proportion of health expenditures in China's GDP is relatively low, so more investment is needed in the healthcare sector, and the burden of individual medical expenses also needs to be continuously reduced. Discussion: In the future, China should further increase its investment in the medical and health sector. Specifically, the government should persist in investing in fundamental medical and health services. Simultaneously, efforts should be made to establish a scientific cost control mechanism for pharmaceuticals and broaden financing channels for healthcare, such as accelerating the development of commercial health insurance.

Temporal variations and trends prediction of water quality during 2010-2019 in the middle Yangtze River, China.

Water quality plays an important role in river habitats. This study revealed the annual and seasonal variations and trend prediction of water quality in the middle Yangtze River after the third impoundment stage of the Three Gorges Reservoir. Multivariate statistical methods including principal component analysis/factor analysis (PCA/FA), Mann-Kendall (M-K) tests, discriminant analysis (DA), rescaled range (R/S) analysis, and the Canadian Council of Ministers of the Environment Water Quality Index (CCME-WQI) were used. Herein, eight water quality constituents including pH, electrical conductivity (EC), chloride (Cl), dissolved oxygen (DO), ammonia nitrogen (NH3N), total phosphorus (TP), water temperature (T), and permanganate index (CODmn) were monthly monitored in the Jiujiang hydrological transaction during 2010-2019. The information of eight water quality constituents, related to salinity, nutrient status, and oxidation reactions efficiency, was extracted. Water quality status remained as fair-good during 2010-2019 based on the results of CCME-WQI, with the seasonal significance ranked as T > DO > Cl > pH > EC > TP > NH3N > CODmn. In the future decade, annual average T was predicted to continue to increase although it might decrease in the wet season. EC was predicted to continue increasing annually especially in the wet season while Cl might decrease. NH3N and TP might maintain a significant decreasing trend in the future wet and dry seasons. DO maintained significantly increasing especially in the future dry seasons, whereas CODmn will continue to decrease annually and seasonally. The continued alkalization trend of waterbody was predicted, which is more significant in the wet season. The results provide helpful references for the ecological protection of the middle Yangtze River.

Did AI get more negative recently?

In this paper, we classify scientific articles in the domain of natural language processing (NLP) and machine learning (ML), as core subfields of artificial intelligence (AI), into whether (i) they extend the current state-of-the-art by the introduction of novel techniques which beat existing models or whether (ii) they mainly criticize the existing state-of-the-art, i.e. that it is deficient with respect to some property (e.g. wrong evaluation, wrong datasets, misleading task specification). We refer to contributions under (i) as having a 'positive stance' and contributions under (ii) as having a 'negative stance' (to related work). We annotate over 1.5 k papers from NLP and ML to train a SciBERT-based model to automatically predict the stance of a paper based on its title and abstract. We then analyse large-scale trends on over 41 k papers from the last approximately 35 years in NLP and ML, finding that papers have become substantially more positive over time, but negative papers also got more negative and we observe considerably more negative papers in recent years. Negative papers are also more influential in terms of citations they receive.

A Bayesian-based classification framework for financial time series trend prediction.

Financial time series have been extensively studied within the past decades; however, the advent of machine learning and deep neural networks opened new horizons to apply supercomputing techniques to extract more insights from the underlying patterns of price data. This paper presents a tri-state labeling approach to classify the underlying patterns in price data into up, down and no-action classes. The introduction of a no-action state in our novel approach alleviates the burden of denoising the dataset as a preprocessing task. The performance of our labeling algorithm is experimented with using machine learning and deep learning models. The framework is augmented by applying the Bayesian optimization technique for the selection of the best tuning values of the hyperparameters. The price trend prediction module generates the required trading signals. The results show that the average annualized Sharpe ratio as the trading performance metric is about 2.823, indicating the framework produces excellent cumulative returns.

Glucose trend prediction model based on improved wavelet transform and gated recurrent unit.

Glucose trend prediction based on continuous glucose monitoring (CGM) data is a crucial step in the implementation of an artificial pancreas (AP). A glucose trend prediction model with high accuracy in real-time can greatly improve the glycemic control effect of the artificial pancreas and effectively prevent the occurrence of hyperglycemia and hypoglycemia. In this paper, we propose an improved wavelet transform threshold denoising algorithm for the non-linearity and non-smoothness of the original CGM data. By quantitatively comparing the mean square error (MSE) and signal-to-noise ratio (SNR) before and after the improvement, we prove that the improved wavelet transform threshold denoising algorithm can reduce the degree of distortion after the smoothing of CGM data and improve the extraction effect of CGM data features at the same time. Based on this finding, we propose a glucose trend prediction model (IWT-GRU) based on the improved wavelet transform threshold denoising algorithm and gated recurrent unit. We compared the root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination ($ {\mathrm{R}}^{2} $) of Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), Support vector regression (SVR), Gated Recurrent Unit (GRU) and IWT-GRU on the original CGM monitoring data of 80 patients for 7 consecutive days with different prediction horizon (PH). The results showed that the IWT-GRU model outperformed the other four models. At PH = 45 min, the RMSE was 0.5537 mmol/L, MAPE was 2.2147%, $ {\mathrm{R}}^{2} $ was 0.989 and the average runtime was only 37.2 seconds. Finally, we analyze the limitations of this study and provide an outlook on the future direction of blood glucose trend prediction.