Search

The polynomial regression algorithm can fit almost all data sequences by adding high-order terms of independent variables, but on long sequences, it can only reflect the development trends of the sequences, and the predicted value is not accurate enough. RNN has great advantages in predicting and fitting time sequence data, especially Long Short-Term Memory (LSTM) network has good performance in predicting long time sequence. This paper respectively adopts polynomial regression and LSTM analysis and predict the effect of epidemic prevention and control policies in various countries to control the epidemic and the economic recession. The experiment results show the Covid-19 epidemics trend prediction algorithm based on LSTM has better prediction effects, and is helpful in practical life.

In this paper, we investigated the impact of foreign direct investment (FDI) and research as well as development (R&D) on industrial carbon dioxide (CO2) emissions reduction. Firstly, a Panel-Corrected Standard Error (PCSE) model was established based on Chinese provincial panel data during period of 2000–2017. Secondly, a Grey Model (GM) was constructed to predict the growth trend of CO2 emissions, FDI and R&D in China from 2018 to 2030. It was deduced that: (1) FDI do bring negative effects on industrial CO2 emissions at the national level. From the perspective of regions, the coefficients of the eastern, central and western regions are all negatively correlated with CO2 emissions and pass the test of 1% significance level except the northeastern China. (2) In terms of R&D, it showed significantly improvement on CO2 emissions at the national and regional levels. Moreover, the efficiency of industrial CO2 emissions was improved by R&D with a rate of 27.2%, while the related value was increased by FDI with a rate of merely 4.3% at the national level. (3) According to the results of GM (1,1), China is expected to achieve zero growth of CO2 emissions by 2030. The growth rate of FDI gradually decreased to less than 5%, while the R&D input will always be at an annual growth rate of more than 5% for a long period of time in the future.

Thanks to advanced technologies, ultra-high-frequency limit order book (LOB) data are now available to data analysts. An LOB contains comprehensive information on all transactions in a market. We u...

The stock market is very volatile and non-stationary and generates huge volumes of data in every second. In this article, the existing machine learning algorithms are analyzed for stock market forecasting and also a new pattern-finding algorithm for forecasting stock trend is developed. Three approaches can be used to solve the problem: fundamental analysis, technical analysis, and the machine learning. Experimental analysis done in this article shows that the machine learning could be useful for investors to make profitable decisions. In order to conduct these processes, a real-time dataset has been obtained from the Indian stock market. This article learns the model from Indian National Stock Exchange (NSE) data obtained from Yahoo API to forecast stock prices and targets to make a profit over time. In this article, two separate algorithms and methodologies are analyzed to forecast stock market trends and iteratively improve the model to achieve higher accuracy. Results are showing that the proposed pattern-based customized algorithm is more accurate (10 to 15%) as compared to other two machine learning techniques, which are also increased as the time window increases.

The file attached to this record is the author's final peer reviewed version. The Publisher's final version can be found by following the DOI link. The small and fluctuating samples of lubricating oil data render the wear trend prediction a challenging task in operation and maintenance management of wind turbine gearboxes. To deal with this problem, this paper puts forward a method to enhance the prediction accuracy and robustness of the grey prediction model by introducing multi-source information into traditional grey models. Multi-source information is applied by creating a mapping sequence according to the sequence to be predicted. The significance of the key parameters in the proposed model was investigated by numerical experiments. Based on the results from the numerical experiments, the effectiveness of the proposed method was demonstrated using lubricating oil data captured from industrial wind turbine gearboxes. A comparative analysis was also conducted with a number of selected other models to illustrate the superiority of the proposed model in dealing with small and fluctuating data. Prediction results show that the proposed model is able to relax the quasi-smooth requirement of data sequence and is much more robust in comparison to exponential regression, linear regression and non-equidistance GM(1,1) models.

In this paper, the convolutional recurrent neural network (ConvLSTM) architecture is proposed to predict individual stock prices. The characteristics of stock data are automatically extracted throu...

Under the background of the big data era, the concept and technology of civil aircraft health management have been continuously innovated. This paper investigates the progress of health management technologies of mainstream manufacturers and airlines. Based on the analysis of operational data, this paper proposes a big data platform architecture for civil health management applications. Besides, the case of applying QAR data for prediction is introduced in the typical fault of air conditioning system, and the effectiveness of AR model in short-term trend prediction is verified.

It is quite extraordinary that trend and colour forecasting for the fashion industry can be traced back to an atelier in Paris in the early nineteenth century. Frenchman Victor Jean-Claude wandered...

The Stock market trend prediction is an efficient medium for investors, public companies and government to invest money by taking into account the profit and risk. The existing studies on the development of stock-based prediction systems rely on data acquired from social media sources (sentiment-based) and secondary data sources (financial-sites). However, the data acquired from such sources is usually sparse in nature. Moreover, the selection of predictor variables is also poor, which ultimately degrades the performance of prediction model. The problems associated with existing approaches can be overcome by proposing an effective prediction model with improved quality of input data and enhanced selection/inclusion of predictor variables. This work presents the results of stock prediction by applying a multiple regression model using R software. The results obtained show that the proposed system achieved a prediction accuracy of 95% on KSE 100-index dataset, 89% on Lucky Cement, 97% on Abbot Company dataset. Furthermore, user-friendly interface is provided to assist individuals and companies to invest or not in a specific stock.

Given a financial time series such as S&P 500, or any historical data in stock markets, how can we obtain useful information from recent transaction data to predict the ups and downs at the next moment? Recent work on this issue shows initial evidence that machine learning techniques are capable of identifying (non-linear) dependency in the stock market price sequences. However, due to the high volatility and non-stationary nature of the stock market, forecasting the trend of a financial time series remains a big challenge. In this paper, we introduced a new method to simplify noisy-filled financial temporal series via sequence reconstruction by leveraging motifs (frequent patterns), and then utilize a convolutional neural network to capture spatial structure of time series. The experimental results show the efficiency of our proposed method in feature learning and outperformance with 4%-7% accuracy improvement compared with the traditional signal process methods and frequency trading patterns modeling approach with deep learning in stock trend prediction.

This study proposes a model for predicting traffic flow and an algorithm to optimize lane allocation in front of an Austrian toll plaza. The traffic prediction model uses local traffic data from the motorway and is based on time series analysis. The prediction is structured into three levels: trend prediction for all calendar days in the year, long-term prediction for the current day and short-term prediction for the next hour. The error of long-term prediction was less than 15% per hour over the whole day. An optimization algorithm for better lane allocation was developed by using a camera based traffic state detection system at the toll plaza. Based on the measured queue lengths per toll gate lane, the algorithm shifts vehicles to lower queued areas in front of the toll plaza. Therefore, the algorithm was able to reduce travel times up to 6% in daily average and queue lengths with more than 100m up to 30% per lane. The prediction model and optimization algorithm are not site-specific and can also be applied on different toll plazas or bottlenecks on motorways (e.g. road works, border crossing, motorway junction etc.).

Since the beginning of COVID-19 (corona virus disease 2019), the Indian government implemented several policies and restrictions to curtail its spread. The timely decisions taken by the government helped in decelerating the spread of COVID-19 to a large extent. Despite these decisions, the pandemic continues to spread. Future predictions about the spread can be helpful for future policy-making, i.e., to plan and control the COVID-19 spread. Further, it is observed throughout the world that asymptomatic corona cases play a major role in the spread of the disease. This motivated us to include such cases for accurate trend prediction. India was chosen for the study as the population and population density is very high for India, resulting in the spread of the disease at high speed. In this paper, the modified SEIRD (susceptible-exposed-infected-recovered-deceased) model is proposed for predicting the trend and peak of COVID-19 in India and its four worst-affected states. The modified SEIRD model is based on the SEIRD model, which also uses an asymptomatic exposed population that is asymptomatic but infectious for the predictions. Further, a deep learning-based long short-term memory (LSTM) model is also used for trend prediction in this paper. Predictions of LSTM are compared with the predictions obtained from the proposed modified SEIRD model for the next 30 days. The epidemiological data up to 6th September 2020 have been used for carrying out predictions in this paper. Different lockdowns imposed by the Indian government have also been used in modeling and analyzing the proposed modified SEIRD model.

K-neighbor algorithm is an effective way to classify things according to their characteristics. Novel Coronavirus epidemic development and response situation of various countries around the world can be classified to simplify the difficulty of epidemic prevention and control and provide useful information for them. Based on the data of WHO and the existing research results at home and abroad, this paper analyzes the evaluation indexes of national epidemic situation, selects representative countries as training samples, and points out the epidemic situation types of corresponding countries. It aims to provide reference for relevant researchers in epidemic prevention and control and trend prediction through global epidemic classification, and hopes to improve relevant technologies in further research, so as to make the classification more scientific and accurate.

In order to solve the problem of low prediction accuracy of traditional financial and economic growth trend prediction methods, this paper proposes a prediction method of financial economic growth trend based on pvar model. According to the influence parameters of financial and economic growth, the pvar model is constructed, and the impulse response function is used to correct the influence of random disturbance in pvar model to reduce the model error. The ant colony algorithm is used to mine the data of financial economy and cluster the data. Input the above data into the pvar model, and the output of the model is the prediction result of the financial and economic growth trend. Experimental results show that the proposed method can effectively correct the model error, and the prediction results are basically consistent with the actual results. Therefore, the prediction performance of the proposed method is good.

The version of record of this article, first published in Lecture Notes in Computer Science, is available online at Publisher’s website: https://doi.org/10.1007/978-3-030-79463-7_47 This paper uses a new convolutional neural network framework to collect data on leading indicators including historical prices and their futures and options, and use arrays as the input map of the CNN framework for stock prices trend prediction. Experiments are then conducted by the stock markets of the United States and Taiwan using historical data, futures and options as data sets to predict the stock prices. After that, genetic algorithm is then utilized to find trading signals. Results showed that the designed model achieves good return of the investments.

Artificial intelligence (AI) techniques have been widely applied to infectious disease outbreak detection and early warning, trend prediction, and public health response modeling and assessment. Such public health surveillance and response tasks of major importance pose unique technical challenges such as data sparsity, lack of positive training samples, difficulty in developing baselines and quantifying the control measures, and interwoven dependencies between spatiotemporal elements and finer-grained risk analyses through contact and social networks. Traditional public health surveillance relies heavily on statistical techniques. Recent years have seen tremendous growth of AI-enabled methods, including but not limited to deep learning–based models, complementing statistical approaches. This chapter aims to provide a systematic review of these recent advances applying AI techniques to address public health surveillance and response challenges.

Nowadays, law enforcement agencies – LEAs – are forced to deal with extreme volumes of data, being in need to analyse from heterogeneous data sources, uncover hidden relationships, trends and patterns of incidents and ultimately reach solid evidence to be used in court. In this chapter, a system is presented that can assist LEA officers in fighting crime, that, following the collection of the primary data, it applies semantic reasoning tools that allow the system to relate pieces of data, based on their inner relationships, and extract new information based on the asserted facts and rules defined by the LEAs. Then, the results derived by the reasoners and the initial data are fused using appropriate tools including a trajectory, a person and an event fusion tool, to be finally visualized by the proposed bipartite graphs. The proposed system is expected to decrease the time required to solve a crime by LEA’s officers.

Stock price prediction is the act of predicting the value of the stock of a particular company in the future to maximize an investor’s profit. In this paper, we propose a machine learning model for stock price prediction. The machine learning model uses LSTM (Long short term memory networks) and Multiple regression algorithms. Along with the machine learning model we look at a couple of important ratios and sentiment analysis which are indicative of whether a stock is overvalued or undervalued. Our model is designed to be particularly helpful for short-term investors for deciding entry and exit points during stock trading.

This research is mainly about the prediction of the price change in the stock market. Instead of daily change, this paper analyzes the trend of price change for weeks by judging turning points. Deep neural networks will be used as the classifier of true and fake golden crosses to judge the growth trend of price change. Most stocks on the sample list have positive profits after simulated trading of 10 years. Based on the results we may conclude that deep neural networks are helpful to assist users positively for stock investment.

Effective prediction of vibration signals can provide an important basis for early warning of performance degradation and failure of pumped storage units. This paper proposes a vibration signal prediction method for pumped storage units based on variational mode decomposition (VMD) and gated recurrent unit neural network (GRU). The VMD-GRU prediction method has a significant improvement in effectiveness and prediction accuracy, which is of great application significance in actual engineering.

Reliable forecasting of crude oil price has received a prodigious attention by both investment companies and governments. Motivated by this issue, this paper seeks to propose a new hybrid forecasting model for crude oil price trend prediction. For this purpose, the crude oil price series is initially decomposed by variational mode decomposition algorithm, and the multi-modal data features are extracted based on the decomposed modes. The volatility of crude oil prices is simultaneously converted into trend symbols through symbolic time series analysis. Machine learning multi-classifier are then trained with multi-modal data features and historical volatility as input and trend symbols as output. The well-trained models are used to predict the trend symbols of West Texas Intermediate crude oil future price. Empirical results demonstrate that the proposed hybrid forecasting model outperforms its counterparts. Among the classifiers used, the hybrid prediction model using support vector machine classifier exhibits superior predictive ability. The accuracy of the proposed model for predicting high volatility of crude oil prices is evidenced to be better than that of low volatility.

In Recent years many forecasting methods have been proposed and implemented for the stock market trend prediction. In this Chapter, the trend analyses of the stock market prediction are presented by using Hidden Markov Model with the one day difference in close value for a particular period. The probability values π gives the trend percentage of the stock prices which is calculated for all the observe sequence and hidden sequences. This chapter helps for decision makers to make decisions in case of uncertainty on the basis of the percentage of probability values obtained from the steady state probability distribution.

In this paper, an exponential smoothing algorithm is proposed to predict the trend of network security. This method uses exponential smoothing algorithm to predict the amount of network attacks and uses Python language to realize the exponential smoothing algorithm model, and obtains the optimal smoothing index value by calculating the standard deviation. The experimental results show that the exponential smoothing algorithm can predict the amount of network attacks in a period of time, the trend of network security changes is basically consistent with the actual data, the deviation of the predicted value is small, and the trend of network security changes can be better reflected.

Stock price forecast is pivotal for various financial and economic institutions and individuals. The aim of this study is to present viable and general approaches that would improve the understanding of forecasting stock market close price of individual stock. This paper explains processes of applying methods including Autoregressive Integrated Moving Average (ARIMA) and Exponential Smoothing (ETS) on the close price data of S&P500 index, but the ticker of the stock can be swapped for forecasting other stocks. In terms of determining the accuracy of the models, we center on the simplest methodology. Of the two models involved in this study, we compare them on the basis of standard deviation. Stock data are obtained from yahoo finance using quantmod package in R studio. Forecasting result shows that the ARIMA model has a better fit with the data and can give a promising general trend prediction compared with existing methods.

The ongoing pandemic of COVID-19 has aroused widespread concern around the world and poses a severe threat to public health worldwide. In this paper, the autoregressive integrated moving average (ARIMA) model was used to predict the epidemic trend of COVID-19 in mainland of China. We collected the cumulative cases, cumulative deaths, and cumulative recovery in mainland of China from January 20 to June 30, 2020, and divided the data into experimental group and test group. The ARIMA model was fitted with the experimental group data, and the optimal model was selected for prediction analysis. The predicted data were compared with the test group. The average relative errors of actual cumulative cases, deaths, recovery and predicted values in each province are between −22.32%−22.66%, −9.52%−0.08%, −8.84%−1.16, the results of the comprehensive experimental group and test group show The error of fitting and prediction is small, the degree of fitting is good, the model supports and is suitable for the prediction of the epidemic situation, which has practical guiding significance for the prevention and control of the epidemic situation.Highlights:We predicted future COVID-19 occurrences in mainland of China based on ARIMA model.We validated the model based on the previous outbreak data with actual data for June, 2020.The measures taken by the government have contained spread of the epidemicThe combination of multiple models may improve the robustness of the model

In the era of big data, deep learning for predicting stock market prices and trends has become even more popular than before. We collected 2 years of data from Chinese stock market and proposed a comprehensive customization of feature engineering and deep learning-based model for predicting price trend of stock markets. The proposed solution is comprehensive as it includes pre-processing of the stock market dataset, utilization of multiple feature engineering techniques, combined with a customized deep learning based system for stock market price trend prediction. We conducted comprehensive evaluations on frequently used machine learning models and conclude that our proposed solution outperforms due to the comprehensive feature engineering that we built. The system achieves overall high accuracy for stock market trend prediction. With the detailed design and evaluation of prediction term lengths, feature engineering, and data pre-processing methods, this work contributes to the stock analysis research community both in the financial and technical domains.

During the Chinese Spring Festival travel rush in 2020, a new type of pneumonia disease, named COVID-19 subsequently broke out in Wuhan, Hubei province, China. The COVID-19 was quickly spreading in China and emerged nearly all over the world. In this chapter, our motivation is to adopt the deep learning techniques to help clinic doctors to diagnose the patients of COVID-19 and predict the trend of COVID-19. To realize our motivation, we on the one hand adopt deep learning techniques to analyse CT images of patients. The transfer learning and data augmentation techniques are adopted for the lacking of samples in our obtained CT image data set. We build a model by designing and training a new deep network to help clinic doctors to make an appropriate diagnose decision. On the other hand, according to the spreading characteristics of COVID-19 and the controlling measures adopted by Chinese government, we propose to modify the classic SEIR (susceptible-exposed-infectious-recovered) model and establish a new SEIR dynamics model with considering the infectiousness of the people in the latent period and the quarantine period. The appropriate parameters of our modified SEIR model are learned by using deep learning techniques. Our proposed deep learning-based diagnosis for COVID-19 can help medicine doctors to make an appropriate diagnostic decision. Our modified SEIR model can effectively predict the transmission trend of COVID-19 and can be used for short-term trend prediction of the epidemic.

Stock trend prediction has always been the focus of research in the field of financial big data. Stock data is complex nonlinear data, while stock price is changing over time. Based on the characteristics of stock data, this paper proposes a financial big data stock trend prediction algorithm based on attention mechanism (STPA). We adopt Bidirectional Gated Recurrent Unit (BGRU) and attention mechanism to capture the long-term dependence of data on time series. The attention mechanism is used to analyze the weight of the impact of data from different time periods on the trend prediction results, thereby reducing the error of stock data change trend prediction and improving the accuracy of trend prediction. We select the daily closing price data of 10 stocks for model training and performance evaluation. Experimental results demonstrate that the proposed method STPA achieves higher precision, recall rate and F1-Score in predicting stock change trends than the other methods. Compared with mainstream methods, STPA improves the precision by 4%, improves recall by 2.5%, and improves F1-Score by 3.2%.

The stock market is a very important element of a country’s economy. Investment in the stock markets is a common trend among the people throughout the world. However, the investments at the stock markets may become very risky if people are not well aware of the trend of stock prices. People may gain money or lose their entire life savings by wrong investments at the stock markets. Previously the world has witnessed many stock market disasters, the crippling economic condition of an entire nation. This can be avoided to some extent if the market can be predicted in advance. Considering this fact, in this paper, a technique is proposed to predict the stock prices of different companies using the concepts of deep learning. Our system’s performance is evaluated by using data from the websites of the Wall Street Journal and Alphavantage. Experimental results show that our system can perform a very good prediction of stock prices.

Aiming at the problem that the traditional trend prediction method has a low accuracy in predicting the state change of the pump source system due to the high working intensity and bad working environment of the aircraft pump source system, this paper designs a GRU network prediction model and studies the state change trend of the aircraft pump source system. Firstly, the acquired data of aircraft pump source system are preprocessed, and the processed data are reconstructed in phase space to construct the training sample set of network parameters. The network is trained with training samples, and the optimal trend prediction model is created by constantly adjusting the number of hidden layer neurons. The optimal trend prediction model is used to obtain the trend prediction results. In order to demonstrate the effectiveness of the GRU trend prediction model, several other prediction methods have been compared with the GRU in this paper. The results show that GRU trend prediction model has higher prediction accuracy than ARMA model and traditional BP network. Compared with the LSTM prediction model which has similar prediction performance, GRU prediction model has more advantages in model training speed and better engineering application value.

Academic citation graphs represent citation relationships between publications across the full range of academic fields. Top cited papers typically reveal future trends in their corresponding domains which is of importance to both researchers and practitioners. Prior citation prediction methods often require initial citation trends to be established and do not take advantage of the recent advancements in graph neural networks (GNNs). We present GNN-based architecture that predicts the top set of papers at the time of publication. For experiments, we curate a set of academic citation graphs for a variety of conferences and show that the proposed model outperforms other classic machine learning models in terms of the F1-score., Comment: Appeared in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2020. 5 pages, 5 figures

Stock trend prediction refers to predicting future price trend of stocks for seeking profit maximum of stock investment. Although it has aroused broad attention in stock markets, it is still a tough task not only because the stock markets are complex and easily volatile but also because real short-term stock data is so limited that existing stock prediction models could be far from perfect, especially for deep neural networks. As a kind of time-series data, the underlying patterns of stock data are easily influenced by any tiny noises. Thus, how to augment limited stock price data is an open problem in stock trend prediction, since most data augmentation schemes adopted in image processing cannot be brutally used here. To this end, we devise a simple yet effective time-sensitive data augmentation method for stock trend prediction. To be specific, we augment data by corrupting high-frequency patterns of original stock price data as well as preserving low-frequency ones in the frame of wavelet transformation. The proposed method is motivated by the fact that low-frequency patterns without noisy corruptions do not hurt the true patterns of stock price data. Besides, a transformation technique is proposed to recognize the importance of the patterns at varied time points, that is, the information is time-sensitive. A series of experiments carried out on a real stock price dataset including 50 corporation stocks verify the efficacy of our data augmentation method.

This paper formulates memory based learning long short-term sequences to forecast the dengue trend. We believe that this is the first work that shows how to use Long Short Term Memory Network to predict/forecast and determine the dengue trend. This problem of dengue trend prediction can be formulated as a temporal time series regression problem. Long short-term memory networks (LSTM) are the special kind of neural networks solve vanishing and exploding gradient problem that traditional neural networks suffered. LSTM networks have the ability to learn the temporal relationship of time series data and can handle the long-term dependency problems. We explore different LSTM regression methodologies and compare and report their accuracies to confirm the efficacy of our proposed method.

Highlights • Since January 23 2020, stringent measures for controlling novel coronavirus epidemics have been enforced and strengthened in mainland China. • Most infected cases have been quarantined or put in suspected class, which had been ignored in existing models. • Results of our model show that the trend of the epidemics mainly depends on quarantined and suspected cases. • It is important to continue enhancing the quarantine and isolation strategy and improving the detection rate in mainland China., Objectives Since January 23, 2020, stringent measures for controlling the novel coronavirus epidemic have been gradually enforced and strengthened in mainland China. The detection and diagnosis have been improved, as well. However, the daily reported cases remaining at a high level make the epidemic trend prediction difficult. Methods Since the traditional SEIR model does not evaluate the effectiveness of control strategies, a novel model in line with the current epidemic's process and control measures was proposed, utilizing multisource datasets including the cumulative number of reported, deceased, quarantined and suspected cases. Results Results show that the trend of the epidemic mainly depends on quarantined and suspected cases. The predicted cumulative numbers of quarantined and suspected cases nearly reached static states, and their inflection points have already been achieved, with the epidemic's peak coming soon. The estimated effective reproduction numbers using model-free and model-based methods are decreasing, as well as new infections, while newly reported cases are increasing. Most infected cases have been quarantined or put in the suspected class, which has been ignored in existing models. Conclusions The uncertainty analyses reveal that the epidemic is still uncertain, and it is important to continue enhancing the quarantine and isolation strategy and improving the detection rate in mainland China.