Backward Prediction Research Articles

An Improved Sequential Recommendation Algorithm based on Short‐Sequence Enhancement and Temporal Self‐Attention Mechanism

Sequential recommendation algorithm can predict the next action of a user by modeling the user’s interaction sequence with an item. However, most sequential recommendation models only consider the absolute positions of items in the sequence, ignoring the time interval information between items, and cannot effectively mine user preference changes. In addition, existing models perform poorly on sparse data sets, which make a poor prediction effect for short sequences. To address the above problems, an improved sequential recommendation algorithm based on short‐sequence enhancement and temporal self‐attention mechanism is proposed in this paper. In the proposed algorithm, a backward prediction model is trained first, to predict the prior items in the user sequence. Then, the reverse prediction model is used to generate a batch of pseudo‐historical items before the initial items of the short sequence, to achieve the goal of enhancing the short sequence. Finally, the absolute position information and time interval information of the user sequence are modeled, and a time‐aware self‐attention model is adopted to predict the user’s next action and generate a recommendation list. Various experiments are conducted on two public data sets. The experimental results show that the method proposed in this paper has excellent performance on both dense and sparse data sets, and its effect is better than that of the state of the art.

Landslide susceptibility evolution in an area affected by the 2015 Nepal earthquakes, derived from multitemporal landslide inventories

The Mw 7.8 and Mw 7.3 Nepal earthquakes in 2015 triggered >47,000 landslides and significantly enhanced landslide activity in the affected region in the following years, likely due to a drop in rock strength of the slopes there. This work analyzed the evolution of landslide susceptibility of an area close to the Zhangmu border crossing on the Araniko Highway in Nepal, which was seriously affected by the mainshock, aftershock, and monsoon rainfalls. Using eight landslide inventories we constructed to register landslides triggered by monsoon rainfalls before 2015, the mainshock, the aftershock, and the monsoon rainfalls in 2015–2019, eight susceptibility assessment models were built using the logistic regression method considering 11 control factors of landslides. The changes in the high-susceptibility zones before, during, and after the earthquakes were examined. The backward prediction abilities of the models were checked by taking landsliding and nonlandsliding data in the following years. The results show that the best susceptibility mapping results are obtained based on landslides triggered by the mainshock and the aftershock, which both have successful prediction rates >87%. The high-susceptibility regions have moved from the slope surfaces to channels since 2015, suggesting that transportation of residual debris induced by the strong ground motion resulted from erosion related to the strong post-seismic rainfall and the rivers. A series of attempts using susceptibility models to backward predict landslides in the following years demonstrate that the combination of the susceptibility results of pre-seismic and mainshock-triggered landslides, which have a higher prediction accuracy, has prediction rates >70% until 2018. This study not only will further the understanding of post-seismic landsliding effects, but also will aid in efforts toward future seismic landslide mitigation and post-seismic reconstruction.

Accelerating Additive Design With Probabilistic Machine Learning

Abstract Additive manufacturing (AM) has been growing rapidly to transform industrial applications. However, the fundamental mechanism of AM has not been fully understood which resulted in low success rate of building. A remedy is to introduce surrogate modeling based on experimental dataset to assist additive design and increase design efficiency. As one of the first papers for predictive modeling of AM especially direct energy deposition (DED), this paper discusses a bidirectional modeling framework and its application to multiple DED benchmark designs including: (1) forward prediction with cross-validation, (2) global sensitivity analyses, (3) backward prediction and optimization, and (4) intelligent data addition. Approximately 1150 mechanical tensile test samples were extracted and tested with input variables from machine parameters, postprocess, and output variables from mechanical, microstructure, and physical properties.

Backward Prediction Confirms the Conclusion on Local Plant Population Viability

Backward Prediction Confirms the Conclusion on Local Plant Population Viability

Prediction of the axial compressive strength of circular concrete-filled steel tube columns using sine cosine algorithm-support vector regression

In this study, support vector regression optimized by a sine cosine algorithm was used to predict the axial bearing capacities of circular concrete-infilled steel tubular columns. A database consisting of 478 test specimens was constructed to train and validate the proposed model, and the effects of the data composition and the sine cosine optimization were investigated. The accuracy of the proposed method was verified by comparing the proposed model’s performance to those of other widely used machine learning methods and design formulae. The stability of the proposed method was verified by evaluating the average model performance of 100 random numerical experiments. For computer aided intelligent design, a backward prediction analysis using machine learning was performed, and the results indicate that the combination of support vector regression and the sine cosine algorithm can provide preliminary design assistance, whereas the empirical formulae cannot.

Reciprocal Twin Networks for Pedestrian Motion Learning and Future Path Prediction

Modeling the moving behaviors and predicting the future paths of pedestrians, especially for those in complex scenes, remain a challenging problem in machine learning. We recognize that human motion trajectories, governed by social norms and constrained by physical structures of the surrounding environment, are both forward predictable and backward predictable. Motivated by this observation, we develop a new approach, called reciprocal twin networks, for human trajectory learning and prediction. We design two networks, a forward prediction network to predict future trajectory from past observations and a backward prediction that performs the trajectory prediction backward in time. The backward prediction network serves as the inverse operation of the forward prediction network, forming a reciprocal constraint. During the training stage, this reciprocal constraint allows them to be jointly learned for accurate and robust human trajectory prediction. During the inference stage, we borrow the concept of adversarial attack of deep neural networks, which iteratively modifies the input of the network to match the given or forced network output, and develop a new method, called reciprocal attack for matched prediction, to achieve accurate human trajectory prediction. Our experimental results on benchmark datasets demonstrate that our new method outperforms the state-of-the-art methods for human trajectory prediction.

Chaos Theory and ARTFA: Emerging Tools for Interpreting ECG Signals to Diagnose Cardiac Arrhythmias

Timely detection of cardiac abnormalities from an Electrocardiogram (ECG) signal is very essential. This requires its appropriate and efficient processing. In the literature, most of the researchers focussed on linear techniques that were applied on filtered ECG datasets leaving an ample scope for exploring the use of non-linear techniques in the presence and absence of natural noise-processes. Therefore, there is a need of supplementing the existing research on ECG signal interpretation by using non-linear techniques on noisy ECG data. Non-linear techniques are expected to yield supplementary clues about the non-linearities in the underlying cardiovascular system. One such promising non-linear technique, known as chaos theory (analysis), has been considered here to estimate reliable and robust thresholds for R-peak detection using fractal dimension, Approximate Entropy, Sample Entropy, correlation dimension, and Lyapunov exponent based on time-delay dimension (embedding). Also, time–frequency analysis techniques have shown their effectiveness for analyzing such types of non-linear and non-stationary signals due to simultaneous interpretation of the signal in both time and frequency domain. Among existing TFA techniques such as wavelet transform, short time Fourier transform, Hilbert transform, Auto-regressive Time Frequency Analysis (ARTFA) offers good time–frequency resolution. Therefore, Chaos theory and ARTFA have been considered in this paper. First, raw ECG signal was filtered using Savitzky Golay Digital Filtering (SGDF) because it retains all important signal features after filtering. Second, a novel optimal trajectory detection step was proposed on the basis of phase space reconstruction (attractors) in chaos theory. Third, ARTFA has been used to find the spectral components of the extracted features using chaos theory. Here, ARTFA has been used for finding the autoregressive coefficients in the first step and time–frequency description in the second step. Burg method has been considered for Auto-regressive modeling to fit an ARTFA model for analyzing ECG signal by minimizing (least squares) the forward and backward prediction errors. MIT-BIH arrhythmia database has been considered for validating the present research effort. Some real time signals were also tested to explore direct usage of the proposed technique in practical applications. The obtained results show that the proposed technique enhances sensitivity, positive predictive value and accuracy thereby improving the detection of arrhythmias. Computational cost of the proposed technique is reduced to a great extent by using the chaos theory (analysis) yielding efficient detection performance. All results have been obtained in MATLAB environment R2011a. Improved values viz. 99.96% SE, 99.97% PPV, and 99.93% ACC are obtained.

The impact of COVID-19 on Farmers' Economic Income in Hubei Province of China

This paper discusses the statistical measurement of the impact of COVID-19 major emergencies on farmers' economic income in Hubei Province. Hubei Province was selected as the object of analysis, and five data of total output value of agriculture, forestry, animal husbandry, fishery and per capita disposable income of farmers in Hubei Province from the first quarter of 2013 to the second quarter of 2020 were collected by using the Internet. Since all the collected data were macroeconomic data, these data were taken the logarithm to meet the economic significance. The per capita disposable income of farmers was taken as the response variable, and the main factors affecting farmers' income were obtained by factor analysis. Livestock husbandry and fishery industries were the main industries in Hubei Province. Then the score of factor analysis were taken as explained variable to establish regression model composed of influencing factors. This paper use the multiple linear regression, support vector regression to fitting and forecasting data, ARIMA model of time series analysis, introduced at the same time, through the AIC model choice, with the first quarter of 2013 to 2019 in the second quarter fitting training, backward prediction two quarters, and three or four quarter of 2019 compared with the real data, through to the predicted results of the sequence diagram and evaluation index model to compare the mean square error (RMSE). Three models predict per capita disposable income of farmers in the first and second quarter of 2020. It has been found that performance better ARIMA model in the model compare is worse than before, and three kinds of predicted values are higher than the real value of the model, showed the outbreak to the influence of the agricultural economy in hubei province is serious. On this basis, taking into account the characteristics of geomorphic climate in Hubei province, the constructive suggestions are put forward.

The impact of COVID-19 on farmers' economic income in Hubei Province of China

This paper discusses the statistical measurement of the impact of COVID-19 majoremergencies on farmers' economic income in Hubei Province. Hubei Province wasselected as the object of analysis, and five data of total output value of agriculture,forestry, animal husbandry, fishery and per capita disposable income of farmers inHubei Province from the first quarter of 2013 to the second quarter of 2020 werecollected by using the Internet. Since all the collected data were macroeconomic data,these data were taken the logarithm to meet the economic significance.The per capita disposable income of farmers was taken as the response variable, andthe main factors affecting farmers' income were obtained by factor analysis.Livestock husbandry and fishery industries were the main industries in HubeiProvince. Then the score of factor analysis were taken as explained variable toestablish regression model composed of influencing factors. This paper use themultiple linear regression, support vector regression to fitting and forecasting data,ARIMA model of time series analysis, introduced at the same time, through the AICmodel choice, with the first quarter of 2013 to 2019 in the second quarter fittingtraining, backward prediction two quarters, and three or four quarter of 2019compared with the real data, through to the predicted results of the sequence diagramand evaluation index model to compare the mean square error (RMSE).Three models predict per capita disposable income of farmers in the first and secondquarter of 2020. It has been found that performance better ARIMA model in themodel compare is worse than before, and three kinds of predicted values are higherthan the real value of the model, showed the outbreak to the influence of theagricultural economy in hubei province is serious.On this basis, taking into accountthe characteristics of geomorphic climate in Hubei province, the constructivesuggestions are put forward.

On farmers’ Economic Income in Hubei Province of China During the COVID-19 Epidemic

This paper discusses the statistical measurement of the impact of COVID-19 major emergencies on farmers’ economic income in Hubei Province. Hubei Province was selected as the object of analysis, and five data of total output value of agriculture, forestry, animal husbandry, fishery and per capita disposable income of farmers in Hubei Province from the first quarter of 2013 to the second quarter of 2020 were collected by using the Internet. Since all the collected data were macroeconomic data, these data were taken the logarithm to meet the economic significance. The per capita disposable income of farmers was taken as the response variable, and the main factors affecting farmers’ income were obtained by factor analysis. Livestock husbandry and fishery industries were the main industries in Hubei Province. Then the score of factor analysis were taken as explained variable to establish a regression model composed of influencing factors. This paper uses the multiple linear regression, support vector regression to fitting and forecasting data, ARIMA model of time series analysis, introduced at the same time, through the AIC model choice, with the first quarter of 2013 to 2019 in the second quarter fitting training, backward prediction two quarters, and three or four quarter of 2019 compared with the real data, through to the predicted results of the sequence diagram and evaluation index model to compare the mean square error (RMSE). Three models predict per capita disposable income of farmers in the first and second quarter of 2020. It has been found that performance better ARIMA model in the model compare is worse than before, and three kinds of predicted values are higher than the real value of the model, showed the outbreak to the influence of the agricultural economy in Hubei province is serious. On this basis, taking into account the characteristics of geomorphic climate in Hubei province, constructive suggestions are put forward.

On universal algorithms for classifying and predicting stationary processes

This is a survey of results on universal algorithms for classification and prediction of stationary processes. The classification problems include discovering the order of a k-step Markov chain, determining memory words in finitarily Markovian processes and estimating the entropy of an unknown process. The prediction problems cover both discrete and real valued processes in a variety of situations. Both the forward and the backward prediction problems are discussed with the emphasis being on pointwise results. This survey is just a teaser. The purpose is merely to call attention to results on classification and prediction. We will refer the interested reader to the sources. Throughout the paper we will give illuminating examples.

New statistical RI index allow to better track the dynamics of COVID-19 outbreak in Italy

COVID-19 pandemic in Italy displayed a spatial distribution that made the tracking of its time course quite difficult. The most relevant anomaly was the marked spatial heterogeneity of COVID-19 diffusion. Lombardia region accounted for around 60% of fatal cases (while hosting 15% of Italian population). Moreover, 86% of fatalities concentrated in four Northern Italy regions. The ‘explosive’ outbreak of COVID-19 in Lombardia at the very beginning of pandemic fatally biased the R-like statistics routinely used to control the disease dynamics. To (at least partially) overcome this bias, we propose a new index RI = dH/dI (daily derivative ratio of H and I, given H = Healed and I = Infected), corresponding to the ratio between healed and infected patients relative daily changes. The proposed index is less flawed than R by the uncertainty related to the estimated number of infected persons and allows to follow (and possibly forecast) epidemic dynamics in a largely model-independent way. To analyze the dynamics of the epidemic, starting from the beginning of the virus spreading—when data are insufficient to make an estimate by adopting SIR model—a "sigmoidal family with delay" logistic model was introduced. That approach allowed in estimating the epidemic peak using the few data gathered even before mid-March. Based on this analysis, the peak was correctly predicted to occur by end of April. Analytical methodology of the dynamics of the epidemic we are proposing herein aims to forecast the time and intensity of the epidemic peak (forward prediction), while allowing identifying the (more likely) beginning of the epidemic (backward prediction). In addition, we established a relationship between hospitalization in intensive care units (ICU) versus deaths daily rates by avoiding the necessity to rely on precise estimates of the infected fraction of the population The joint evolution of the above parameters over time allows for a trustworthy and unbiased estimation of the dynamics of the epidemic, allowing us to clearly detect the qualitatively different character of the ‘so-called’ second wave with respect to the previous epidemic peak.

Epilepsy prediction through optimized multidimensional sample entropy and Bi-LSTM

ObjectiveEpilepsy is a repetitive and transient brain dysfunction caused by abnormal discharge of brain neurons. Sudden epileptic seizures may affect the daily life of patients. Therefore, real-time monitoring and prediction of epilepsy has important clinical meaning. MethodsIn this paper, the characteristics of M-SampEn were extracted from 23 EEG signals and M-SampEn was specifically optimized to enhance efficiency. Then the Bi-LSTM may predict the trend of M-SampEn. The predicted M-SampEn was classified to determine if an epileptic seizure is imminent. ResultsComparing the classification accuracy, sensitivity, specificity and PPV of SampEn and M-SampEn, M-SampEn is found to have better performance. The prediction time is 5 minutes. The results demonstrate an accuracy of 80.09% and a FPR of 0.26/h for epileptic seizure prediction. Comparison with existing method(s)The optimized multidimensional sample entropy presented in this paper is more able to distinguish between the normal state and ictal of epilepsy. This paper also proposes a backward prediction method that is different from traditional epileptic seizure prediction. ConclusionsThe research provides a high comprehensive performance epileptic prediction method with a F1 score of 0.83. The accuracy of 80.09% and the FPR of 0.26/h prove that the proposed method is able to predict seizures.

A Quadruple Diffusion Convolutional Recurrent Network for Human Motion Prediction

Recurrent neural network (RNN) has become popular for human motion prediction thanks to its ability to capture temporal dependencies. However, it has limited capacity in modeling the complex spatial relationship in the human skeletal structure. In this work, we present a novel diffusion convolutional recurrent predictor for spatial and temporal movement forecasting, with multi-step random walks traversing bidirectionally along an adaptive graph to model interdependency among body joints. In the temporal domain, existing methods rely on a single forward predictor with the produced motion deflecting to the drift route, which leads to error accumulations over time. We propose to supplement the forward predictor with a forward discriminator to alleviate such motion drift in the long term under adversarial training. The solution is further enhanced by a backward predictor and a backward discriminator to effectively reduce the error, such that the system can also look into the past to improve the prediction at early frames. The two-way spatial diffusion convolutions and two-way temporal predictors together form a quadruple network. Furthermore, we train our framework by modeling the velocity from observed motion dynamics instead of static poses to predict future movements that effectively reduces the discontinuity problem at early prediction. Our method outperforms the state of the arts on both 3D and 2D datasets, including the Human3.6M, CMU Motion Capture and Penn Action datasets. The results also show that our method correctly predicts both high-dynamic and low-dynamic moving trends with less motion drift.

Occurrence of backward bifurcation and prediction of disease transmission with imperfect lockdown: A case study on COVID-19.

The outbreak of COVID-19 caused by SARS-CoV-2 is spreading rapidly around the world, which is causing a major public health concerns. The outbreaks started in India on March 2, 2020. As of April 30, 2020, 34,864 confirmed cases and 1154 deaths are reported in India and more than 30,90,445 confirmed cases and 2,17,769 deaths are reported worldwide. Mathematical models may help to explore the transmission dynamics, prediction and control of COVID-19 in the absence of an appropriate medication or vaccine. In this study, we consider a mathematical model on COVID-19 transmission with the imperfect lockdown effect. The basic reproduction number, R 0, is calculated using the next generation matrix method. The system has a disease-free equilibrium (DFE) which is locally asymptotically stable whenever R 0 < 1. Moreover, the model exhibits the backward bifurcation phenomenon, where the stable DFE coexists with a stable endemic equilibrium when R 0 < 1. The epidemiological implications of this phenomenon is that the classical epidemiological requirement of making R 0 less than unity is only a necessary, but not sufficient for effectively controlling the spread of COVID-19 outbreak. It is observed that the system undergoes backward bifurcation which is a new observation for COVID-19 disease transmission model. We also noticed that under the perfect lockdown scenario, there is no possibility of having backward bifurcation. Using Lyapunov function theory and LaSalle Invariance Principle, the DFE is shown globally asymptotically stable for perfect lockdown model. We have calibrated our proposed model parameters to fit daily data from India, Mexico, South Africa and Argentina. We have provided a short-term prediction for India, Mexico, South Africa and Argentina of future cases of COVID-19. We calculate the basic reproduction number from the estimated parameters. We further assess the impact of lockdown during the outbreak. Furthermore, we find that effective lockdown is very necessary to reduce the burden of diseases.

Anomaly detection in surveillance video based on bidirectional prediction

With the development of information technology and the popularization of monitoring network, how to quickly and automatically detect abnormal behaviors in surveillance video is becoming more and more important for public security and smart city. The emergence of deep learning has greatly promoted the development of anomaly detection and much remarkable work has been presented on this topic. However, the existing approaches for anomaly detection generally encounter problems such as insufficient utilization of motion patterns and instability on different datasets. To improve the performance of anomaly detection in surveillance video, we propose a framework based on bidirectional prediction, which predicts the same target frame by the forward and the backward prediction subnetworks, respectively. Then the loss function is constructed based on the real target frame and its bidirectional prediction frame. Furthermore, we also propose an anomaly score estimation method based on the sliding window scheme which focuses on the foregrounds of the prediction error map. The comparison with the state-of-the-art shows that the proposed model outperforms most competing models on different video surveillance datasets.

Welding parameters prediction for arbitrary layer height in robotic wire and arc additive manufacturing

In wire and arc additive manufacturing, the weld bead geometry determined the slicing layer height, which was decided by the welding parameters. Generally, the determination of the welding parameters relied on empirical and experimental data through the trial-and-error methods that incur considerable time and cost. To obtain the proper welding process parameters according to the desired single bead geometry and layer height, a full factorial experimental design matrix was applied to collect the original data of welding parameters and bead geometrical variables. A forward artificial neural network (FANN) was built to predict the bead geometry form the welding parameters. Then, a closed-loop iteration method combined a genetic algorithm (GA) and the FANN model (FANN-GA) was developed to search for the most optimal welding process parameters in accordance with the selected bead geometrical variables. The results confirmed that the FANN-GA model has a good performance on the backward prediction of the welding process parameters compared with the direct backward artificial neural network (BANN). Several groups of single layer multi-bead and multi-layer multi-bead experiment were performed to testify the proposed method, and the relative error between the desired and actual layer height was small. The proposed method makes it possible to fabricate the component with an arbitrary desired layer height, and could be used in the adaptive slicing additive manufacturing or surface coating.

A Novel Deep Learning-Based Spatiotemporal Fusion Method for Combining Satellite Images with Different Resolutions Using a Two-Stream Convolutional Neural Network

Spatiotemporal fusion is considered a feasible and cost-effective way to solve the trade-off between the spatial and temporal resolution of satellite sensors. Recently proposed learning-based spatiotemporal fusion methods can address the prediction of both phenological and land-cover change. In this paper, we propose a novel deep learning-based spatiotemporal data fusion method that uses a two-stream convolutional neural network. The method combines both forward and backward prediction to generate a target fine image, where temporal change-based and a spatial information-based mapping are simultaneously formed, addressing the prediction of both phenological and land-cover changes with better generalization ability and robustness. Comparative experimental results for the test datasets with phenological and land-cover changes verified the effectiveness of our method. Compared to existing learning-based spatiotemporal fusion methods, our method is more effective in predicting phenological change and directly reconstructing the prediction with complete spatial details without the need for auxiliary modulation.

Appraising the Financial Sustainability of a Pension System with Signal Processing

One key issue of the Spanish pension system is its financial sustainability in regard to slumping fertility rate and rising longevity of the Spanish population. The paper presents a versatile and robust model that may help pension managers gain insight into the future Spanish pyramid of ages, from which they will appraise cash inflows and outflows of the pension system. The model forecasts ninety years of the Spanish population for each cohort of the pyramid of ages. Borrowed from the signal processing discipline, the model relies on the Burg method which fits a pth order autoregressive (AR) model to the input signal by minimizing (least squares) the forward and backward prediction errors while constrai-ning the AR parameters to satisfy the Levinson-Durbin recursión, then uses an infinite impulse response prediction error filter. Results add better perspective and insight to the Spanish population projection forecasted by the United Nations Population Division.

Complexity reduction method for High Efficiency Video Coding encoding based on scene-change detection and image texture information

Recently, as demand for high-quality video and realistic media has increased, High Efficiency Video Coding has been standardized. However, High Efficiency Video Coding requires heavy cost in terms of computational complexity to achieve high coding efficiency, which causes problems in fast coding processing and real-time processing. In particular, High Efficiency Video Coding inter-coding has heavy computational complexity, and the High Efficiency Video Coding inter prediction uses reference pictures to improve coding efficiency. The reference pictures are typically signaled in two independent lists according to the display order, to be used for forward and backward prediction. If an event occurs in the input video, such as a scene change, the inter prediction performs unnecessary computations. Therefore, the reference picture list should be reconfigured to improve the inter prediction performance and reduce computational complexity. To address this problem, this article proposes a method to reduce computational complexity for fast High Efficiency Video Coding encoding using information such as scene changes obtained from the input video through preprocessing. Furthermore, reference picture lists are reconstructed by sorting the reference pictures by similarity to the current coded picture using Angular Second Moment, Contrast, Entropy, and Correlation, which are image texture parameters from the input video. Simulations are used to show that both the encoding time and coding efficiency could be improved simultaneously by applying the proposed algorithms.