Time Window Size Research Articles

Decoding Articulation Motor Imagery using Early Connectivity Information in the Motor Cortex: A Functional Near-infrared Spectroscopy Study.

Brain computer interface (BCI) based on speech imagery can help people with motor disorders communicate their thoughts to the outside world in a natural way. Due to being portable, non-invasive, and safe, functional near-infrared spectroscopy (fNIRS) is preferred for developing BCIs. Previous BCIs based on fNIRS mainly relied on activation information, which ignored the functional connectivity between neural areas. In this study, a 4-class speech imagery BCI based on fNIRS is presented to decode simplified articulation motor imagery (only the movements of jaw and lip were retained) of different vowels. Synchronization information in the motor cortex was extracted as features. In multiclass (four classes) settings, the mean subject-dependent classification accuracies approximated or exceeded 40% in the 0-2.5 s and 0-10 s time windows, respectively. In binary class settings (the average classification accuracies of all pairwise comparisons between two vowels), the mean subject-dependent classification accuracies exceeded 70% in the 0-2.5 s and 0-10 s time windows. These results demonstrate that connectivity features can effectively differentiate different vowels even if the time window size was reduced from 10 s to 2.5 s and the decoding performance in both the time windows was almost the same. This finding suggests that speech imagery BCI based on fNIRS can be further optimized in terms of feature extraction and command generation time reduction. In addition, simplified articulation motor imagery of vowels can be distinguished, and therefore, the potential contribution of articulation motor imagery information extracted from the motor cortex should be emphasized in speech imagery BCI based on fNIRS to improve decoding performance.

A social media-based framework for quantifying temporal changes to wildlife viewing intensity

Documenting how human pressure on wildlife changes over time is important to minimise potential adverse effects through implementing appropriate management and policy actions; however, obtaining objective measures of these changes and their potential impacts is often logistically challenging, particularly in the natural environment. Here, we developed a modular stochastic model that infers the ratio of actual viewing pressure on wildlife in consecutive time periods (years) using social media, as this medium is widespread and easily accessible. Pressure was calculated from the number of times individual animals appeared in social media in pre-defined time windows, accounting for time-dependent variables that influence them (e.g. number of people with access to social media). Formulas for the confidence intervals of viewing pressure ratios were rigorously developed and validated, and corresponding uncertainty was quantified. We applied the developed framework to calculate changes to wildlife viewing pressure on loggerhead sea turtles (Caretta caretta) at Zakynthos island (Greece) before and during the COVID-19 pandemic (2019–2021) based on 2646 social media entries. Our model ensured temporal comparability across years of social media data grouped in time window sizes, by correcting for the interannual increase of social media use. Optimal sizes for these windows were delineated, reducing uncertainty while maintaining high time-scale resolution. The optimal time window was around 7-days during the peak tourist season when more data were available in all three years, and >15 days during the low season. In contrast, raw social media data exhibited clear bias when quantifying changes to viewing pressure, with unknown uncertainty. The framework developed here allows widely-available social media data to be used objectively when quantifying temporal changes to wildlife viewing pressure. Its modularity allowed viewing pressure to be quantified for all data combined, or subsets of data (different groups, situations or locations), and could be applied to any site supporting wildlife exposed to tourism.

Decoding lexical tones and vowels in imagined tonal monosyllables using fNIRS signals

Objective. Speech is a common way of communication. Decoding verbal intent could provide a naturalistic communication way for people with severe motor disabilities. Active brain computer interaction (BCI) speller is one of the most commonly used speech BCIs. To reduce the spelling time of Chinese words, identifying vowels and tones that are embedded in imagined Chinese words is essential. Functional near-infrared spectroscopy (fNIRS) has been widely used in BCI because it is portable, non-invasive, safe, low cost, and has a relatively high spatial resolution. Approach. In this study, an active BCI speller based on fNIRS is presented by covertly rehearsing tonal monosyllables with vowels (i.e. /a/, /i/, /o/, and /u/) and four lexical tones in Mandarin Chinese (i.e. tones 1, 2, 3, and 4) for 10 s. Main results. fNIRS results showed significant differences in the right superior temporal gyrus between imagined vowels with tone 2/3/4 and those with tone 1 (i.e. more activations and stronger connections to other brain regions for imagined vowels with tones 2/3/4 than for those with tone 1). Speech-related areas for tone imagery (i.e. the right hemisphere) provided majority of information for identifying tones, while the left hemisphere had advantages in vowel identification. Having decoded both vowels and tones during the post-stimulus 15 s period, the average classification accuracies exceeded 40% and 70% in multiclass (i.e. four classes) and binary settings, respectively. To spell words more quickly, the time window size for decoding was reduced from 15 s to 2.5 s while the classification accuracies were not significantly reduced. Significance. For the first time, this work demonstrated the possibility of discriminating lexical tones and vowels in imagined tonal syllables simultaneously. In addition, the reduced time window for decoding indicated that the spelling time of Chinese words could be significantly reduced in the fNIRS-based BCIs.

SaARSP: An Architecture for Systolic-Array Acceleration of Recurrent Spiking Neural Networks

Spiking neural networks (SNNs) are brain-inspired event-driven models of computation with promising ultra-low energy dissipation. Rich network dynamics emergent in recurrent spiking neural networks (R-SNNs) can form temporally based memory, offering great potential in processing complex spatiotemporal data. However, recurrence in network connectivity produces tightly coupled data dependency in both space and time, rendering hardware acceleration of R-SNNs challenging. We present the first work to exploit spatiotemporal parallelisms to accelerate the R-SNN-based inference on systolic arrays using an architecture called SaARSP. We decouple the processing of feedforward synaptic connections from that of recurrent connections to allow for the exploitation of parallelisms across multiple time points. We propose a novel time window size optimization (TWSO) technique, to further explore the temporal granularity of the proposed decoupling in terms of optimal time window size and reconfiguration of the systolic array considering layer-dependent connectivity to boost performance. Stationary dataflow and time window size are jointly optimized to trade off between weight data reuse and movements of partial sums, the two bottlenecks in latency and energy dissipation of the accelerator. The proposed systolic-array architecture offers a unifying solution to an acceleration of both feedforward and recurrent SNNs, and delivers 4,000X EDP improvement on average for different R-SNN benchmarks over a conventional baseline.

Dynamic analysis and community recognition of stock price based on a complex network perspective

In the context of the 2008 financial crisis, this paper focuses on the Chinese A-share stock prices in ten years (from January 1, 2003 to December 31, 2012) to explore some possible cautionary phenomena based on a complex network perspective. Firstly, with the help of the threshold method and taking monthly and fixed time (20 days, 40 days and 60 days) as time windows, this paper establishes highly correlated networks of daily opening price and backward answer authority closing price respectively. Notice that the correlation between stock price and its category in Citic One-level industry index is misty. And then, it is interesting to discover some special behaviors of stock prices by the evolution of different network features and community identification, so as to provide guidance for crisis identification and control. The results show that the size of time windows has a certain influence on the evolution, and network parameters appear obvious particularity near the 2008 crisis for the networks with fixed times of monthly and 20 days. For example, the average path length reaches a maximum around May 2009; the average degree is low and stable around May 2005 to December 2006 but fluctuates greatly around May 2008 to September 2008; the clustering coefficient fluctuates greatly and reaches the maximum value 0.78 during May 2008 to August 2008; the assortativity coefficient reaches its maximum value at a certain time and fluctuates relatively greatly. In addition, the community identification results indicate that near the crisis, cluster communities with high network density, which contain a relatively large number of nodes, appear.

Problem Decomposition and Multi-shot ASP Solving for Job-shop Scheduling

AbstractScheduling methods are important for effective production and logistics management, where tasks need to be allocated and performed with limited resources. In particular, the Job-shop Scheduling Problem (JSP) is a well known and challenging combinatorial optimization problem in which tasks sharing a machine are to be arranged in a sequence such that encompassing jobs can be completed as early as possible. Given that already moderately sized JSP instances can be highly combinatorial, and neither optimal schedules nor the runtime to termination of complete optimization methods is known, efficient approaches to approximate good-quality schedules are of interest. In this paper, we propose problem decomposition into time windows whose operations can be successively scheduled and optimized by means of multi-shot Answer Set Programming (ASP) solving. From a computational perspective, decomposition aims to split highly complex scheduling tasks into better manageable subproblems with a balanced number of operations so that good-quality or even optimal partial solutions can be reliably found in a small fraction of runtime. Regarding the feasibility and quality of solutions, problem decomposition must respect the precedence of operations within their jobs and partial schedules optimized by time windows should yield better global solutions than obtainable in similar runtime on the entire instance. We devise and investigate a variety of decomposition strategies in terms of the number and size of time windows as well as heuristics for choosing their operations. Moreover, we incorporate time window overlapping and compression techniques into the iterative scheduling process to counteract window-wise optimization limitations restricted to partial schedules. Our experiments on JSP benchmark sets of several sizes show that successive optimization by multi-shot ASP solving leads to substantially better schedules within the runtime limit than global optimization on the full problem, where the gap increases with the number of operations to schedule. While the obtained solution quality still remains behind a state-of-the-art Constraint Programming system, our multi-shot solving approach comes closer the larger the instance size, demonstrating good scalability by problem decomposition.

Identification and Suppression of Magnetotelluric Noise via a Deep Residual Network

The magnetotelluric (MT) method is widely applied in petroleum, mining, and deep Earth structure exploration but suffers from cultural noise. This noise will distort apparent resistivity and phase, leading to false geological interpretation. Therefore, denoising is indispensable for MT signal processing. The sparse representation method acts as a critical role in MT denoising. However, this method depends on the sparse assumption leading to inadequate denoising results in some cases. We propose an alternative MT denoising approach, which can achieve accurate denoising without assumptions on datasets. We first design a residual network (ResNet), which has an excellent fitting ability owing to its deep architecture. In addition, the ResNet network contains skip-connection blocks to guarantee the robustness of network degradation. As for the number of training, validation, and test datasets, we use 10,000,000; 10,000; and 100 field data, respectively, and apply the gradual shrinkage learning rate to ensure the ResNet’s generalization. In the noise identification stage, we use a small-time window to scan the MT time series, after which the gramian angular field (GAF) is applied to help identify noise and divide the MT time series into noise-free and noise data. We keep the noise-free data section in the denoising stage, and the noise data section is fed into our network. In our experiments, we test the performances of different time window sizes for noise identification and suppression and record corresponding time consumption. Then, we compare our approach with sparse representation methods. Testing results show that our approach can obtain the desired denoising results. The accuracy and loss curves show that our approach can well suppress the MT noise, and our network has a good generalization. To further validate our approach’s effectiveness, we show the apparent resistivity, phase, and polarization direction of test datasets. Our approach can adjust the distortion of apparent resistivity and phase and randomize the polarization direction distribution. Although our approach requires the high quality of the training dataset, it achieves accurate MT denoising after training and can be meaningful in cases of a severe MT noisy environment.

CORE

Complex Event Recognition (CER) systems are a prominent technology for finding user-defined query patterns over large data streams in real time. CER query evaluation is known to be computationally challenging, since it requires maintaining a set of partial matches, and this set quickly grows super-linearly in the number of processed events. We present CORE, a novel COmplex event Recognition Engine that focuses on the efficient evaluation of a large class of complex event queries, including time windows as well as the partition-by event correlation operator. This engine uses a novel automaton-based evaluation algorithm that circumvents the super-linear partial match problem: under data complexity, it takes constant time per input event to maintain a data structure that compactly represents the set of partial matches and, once a match is found, the query results may be enumerated from the data structure with output-linear delay. We experimentally compare CORE against state-of-the-art CER systems on real-world data. We show that (1) CORE's performance is stable with respect to both query and time window size, and (2) CORE outperforms the other systems by up to five orders of magnitude on different workloads.

Exploring Eye Movement Biometrics in Real-World Activities: A Case Study of Wayfinding.

Eye movement biometrics can enable continuous verification for highly secure environments such as financial transactions and defense establishments, as well as a more personalized and tailored experience in gaze-based human–computer interactions. However, there are numerous challenges to recognizing people in real environments using eye movements, such as implicity and stimulus independence. In the instance of wayfinding, this research intends to investigate implicit and stimulus-independent eye movement biometrics in real-world situations. We collected 39 subjects’ eye movement data from real-world wayfinding experiments and derived five sets of eye movement features (the basic statistical, pupillary response, fixation density, fixation semantic and saccade encoding features). We adopted a random forest and performed biometric recognition for both identification and verification scenarios. The best accuracy we obtained in the identification scenario was 78% (equal error rate, EER = 6.3%) with the 10-fold classification and 64% (EER = 12.1%) with the leave-one-route-out classification. The best accuracy we achieved in the verification scenario was 89% (EER = 9.1%). Additionally, we tested performance across the 5 feature sets and 20 time window sizes. The results showed that the verification accuracy was insensitive to the increase in the time window size. These findings are the first indication of the viability of performing implicit and stimulus-independent biometric recognition in real-world settings using wearable eye tracking.

Creating grocery delivery hubs for food deserts at local convenience stores via spatial and temporal consolidation

For many socioeconomically disadvantaged customers living in food deserts, the high costs and minimum order size requirements make attended grocery deliveries financially non-viable, although it has a potential to provide healthy foods to the food insecure population. This paper proposes consolidating customer orders and delivering to a neighborhood convenience store instead of home delivery. We employ an optimization framework involving the minimum cost set covering and the capacitated vehicle routing problems. Our experimental studies in three counties in the U.S. suggest that by spatial and temporal consolidation of orders, the deliverer can remove minimum order-size requirements and reduce the delivery costs, depending on various factors, compared to attended home-delivery. We find the number and size of time windows for home delivery to be the most important factor in achieving temporal consolidation benefits. Other significant factors in achieving spatial consolidation include the capacity of delivery vehicles, the number of depots, and the number of customer orders. We also find that the number of partner convenience stores and the walkable distance parameter of the model have a significant impact on the number of accepted orders, i.e., the service level provided by the deliverer. The findings of this study imply consolidated grocery delivery as a viable solution to improve fresh food access in food deserts. In light of the recent global pandemic and its exacerbating effects on food insecurity, the innovative solution proposed in this paper is even more relevant and timely.

Temporal dynamics of clinical risk predictors for hospital-acquired acute kidney injury under different forecast time windows

Massive electronic medical records provide unique opportunities and possibilities for gaining medical knowledge and actionable insights to transform healthcare. Acute kidney injury (AKI) is considered one of the most common complications of acute illness, with a substantial impact on patient outcome and hospital costs. In this study, we developed a machine-learning-based knowledge mining approach, combining eXtreme Gradient Boosting (XGBoost) model and SHapley Additive exPlanations (SHAP) method, to explore the temporal dynamics of clinical risk predictors for hospital-acquired AKI under different forecast time windows. AKI occurred in 7,259 (9.43%) of 76,957 hospital admissions from November 2007 to December 2016 extracted from our institution’s de-identified clinical database. Through qualitative visualization and quantitative analysis, we found and confirmed that relative importance of optimal risk factors fluctuates with the size of AKI prediction time window. For example, the contribution of most time-varying variables (such as medication and lab tests) to AKI risk increases as AKI onset approaches, while most non-time-varying variables (such as age and admission diagnosis) decrease. Besides, the optimal set of AKI risk predictors will also change under different time window. This study provides several practical implications, including recognizing the existence of feature weight volatility as it is desirable for model accuracy, identifying important AKI risk factor for further investigation, and facilitating early accurate prediction of AKI.

Contact Force Detection of Grinding Process Using Frequency Information and Differential Feature on Force Signal

In grinding tasks, the contact force has a significant impact on product surface quality. Therefore, force-sensing technology to detect contact force is important. Although force sensors are widely used for contact force detection, the response of the force sensor includes sensor-specific errors such as offset. In this paper, we propose a contact force detection method based on the combination of frequency information and the differential feature (Δ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> ) of the force signal. The use of high-frequency information reduces the influence of force sensor-specific errors. However, contact force detection using only high-frequency information causes a time delay in the detected value relative to the measured value depending to the frame size of time window used for frequency analysis. To reduce the time delay, high-frequency information and Δ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> are integrated by inputting them into an long short-term memory (LSTM)-based force detection model. To verify the effectiveness of the proposed method, we compared it with a force detection model based on an FNN and CNN on a dataset of plane grinding tasks. Consequently, the detection accuracy of the LSTM-based model was superior to that of the FNN and CNN models. Compared to the LSTM model using only high-frequency information as input, the detection accuracy was 26% higher when the error was small and 57% higher when the error was large. In addition, the time delay was reduced from 166 ms to 30 ms using Δ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> as the input. The frequency information and Δ <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</i> are features calculated from the same force information dataset; therefore, no additional dataset is required.

A novel method of detecting carbon asset price jump characteristics based on significant information shocks

To identify the jump forms of carbon asset price under the shocks of significant information, this paper focuses on two aspects: different jump directions and jump time points. Therefore, a method for simultaneously optimizing time window size and threshold of the LM jump test is proposed. This optimized method has reduced the omission of jump points and improves the effectiveness of the model. Results show that the shocks of significant information cause large magnitude and asymmetric successive jumps of carbon asset price. A downward high-frequency jump indicates that the downward risk of carbon market is significant, and policy events have the biggest impact on the carbon market, which leads to the biggest jump magnitude. This research helps improve the efficiency of carbon pricing, provides practical guidance for decision process of enterprises, and prevents risk in a timely manner.

Idle-state detection in motor imagery of articulation using early information: A functional Near-infrared spectroscopy study

A speech imagery-based brain computer interface (BCI) provides an alternative way for people to interact with the outside world intuitively. Most speech imagery BCIs based on functional near-infrared spectroscopy (fNIRS) are disadvantageous in applications outside the laboratory due to being incapable of detecting asynchronous (self-paced) actions. This work aimed to develop a two-class asynchronous BCI that detects idle state in the context of motor imagery of articulation (MIA). In this study, 19 healthy subjects were asked to rehearse the Chinese vowels/a/and/u/covertly and to accommodate a rest state. A feature selection strategy was designed to combine time domain, spatial domain, and functional connectivity features. All differentiating information was extracted from fNIRS signals in a 0–2.5 s time window. Among single-modality features, the centrality features of brain networks performed better than any other features and yielded a subject-dependent classification accuracy rate of 75.1%. The subject-dependent classification accuracy rate of the combined features at 78.9% was significantly better than that of any single feature. These results demonstrate that it is feasible to detect idle states from active MIA states in a reduced time window size. The proposed combination of information propagation pattern, spatial pattern, and activation pattern was meaningful for extracting discriminative information, hence improving classification accuracy. The high classification accuracy and fast information transfer rate of the presented BCI show promising practicality for real world applications.

DeFFusion: CNN-based Continuous Authentication Using Deep Feature Fusion

Smartphones have become crucial and important in our daily life, but the security and privacy issues have been major concerns of smartphone users. In this article, we present DeFFusion, a CNN-based continuous authentication system using Deep Feature Fusion for smartphone users by leveraging the accelerometer and gyroscope ubiquitously built into smartphones. With the collected data, DeFFusion first converts the time domain data into frequency domain data using the fast Fourier transform and then inputs both of them into a designed CNN, respectively. With the CNN-extracted features, DeFFusion conducts the feature selection utilizing factor analysis and exploits balanced feature concatenation to fuse these deep features. Based on the one-class SVM classifier, DeFFusion authenticates current users as a legitimate user or an impostor. We evaluate the authentication performance of DeFFusion in terms of impact of training data size and time window size, accuracy comparison on different features over different classifiers and on different classifiers with the same CNN-extracted features, accuracy on unseen users, time efficiency, and comparison with representative authentication methods. The experimental results demonstrate that DeFFusion has the best accuracy by achieving the mean equal error rate of 1.00% in a 5-second time window size.

Prediction of PM2.5 Concentration Based on the LSTM-TSLightGBM Variable Weight Combination Model

PM2.5 is one of the main pollutants that cause air pollution, and high concentrations of PM2.5 seriously threaten human health. Therefore, an accurate prediction of PM2.5 concentration has great practical significance for air quality detection, air pollution restoration, and human health. This paper uses the historical air quality concentration data and meteorological data of the Beijing Olympic Sports Center as the research object. This paper establishes a long short-term memory (LSTM) model with a time window size of 12, establishes a T-shape light gradient boosting machine (TSLightGBM) model that uses all information in the time window as the next period of prediction input, and establishes a LSTM-TSLightGBM model pair based on an optimal weighted combination method. PM2.5 hourly concentration is predicted. The prediction results on the test set show that the mean squared error (MAE), root mean squared error (RMSE), and symmetric mean absolute percentage error (SMAPE) of the LSTM-TSLightGBM model are 11.873, 22.516, and 19.540%, respectively. Compared with LSTM, TSLightGBM, the recurrent neural network (RNN), and other models, the LSTM-TSLightGBM model has a lower MAE, RMSE, and SMAPE, and higher prediction accuracy for PM2.5 and better goodness-of-fit.

Effectiveness of cascading time series models based on meteorological factors in improving health risk prediction.

Meteorological factors, which are periodic and regular in a long run, have an unignorable impact on human health. Accurate health risk prediction based on meteorological factors is essential for optimal allocation of resource in healthcare units. However, due to the non-stationary and non-linear nature of the original hospitalization sequence, traditional methods are less robust in predicting it. This study aims to investigate hospital admission prediction models using time series pre-processing algorithms and deep learning approach based on meteorological factors. Using the electronic medical record data from Panyu Central Hospital and meteorological data of Panyu district from 2003 to 2019, 46,089 eligible patients with lower respiratory tract infections (LRTIs) and four meteorological factors were identified to build and evaluate the prediction models. A novel hybrid model, Cascade GAM-CEEMDAN-LSTM Model (CGCLM), was established in combination with generalized additive model (GAM), complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), and long-short term memory (LSTM) networks for predicting daily admissions of patients with LRTIs. The experimental results show that CGCLM multistep method proposed in this paper outperforms single LSTM model in the prediction of health risk time series at different time window sizes. Moreover, our results also indicate that CGCLM has the best prediction performance when the time window is set to 61 days (RMSE = 1.12, MAE = 0.87, R2 = 0.93). Adequate extraction of exposure-response relationships between meteorological factors and diseases and suitable handling of sequence pre-processing have an important role in time series prediction. This hybrid climate-based model for predicting LRTIs disease can also be extended to time series prediction of other epidemic disease.

Missing data imputation using mixture factor analysis for building electric load data

We propose a mixture factor analysis (MFA) method for estimating missing values in building electric load data. Buildings consume a tremendous amount of energy. Thanks to the recent advances in data technologies such as machine learning and applied statistics, data-driven approaches to making buildings more energy-efficient become a major research area. However, building electric load data suffer from quality issues due to data missing originated from malfunctioning sensors, network stability, and other environmental causes. We note that data missing can occur even under advanced Internet Technology (IT) systems such as energy information systems (EIS) and energy management systems (EMS) due to signal stability and low-speed computers. The existence of missing data may significantly affect building operations, causing inaccuracy in evaluating building status and forecasting future electric demands. In this respect, dealing with missing data problems should be as important as developing highly accurate forecasting algorithms. While investigating load data, we find that building electric loads exhibit distinct patterns with cyclic rotations that we can take advantage of in both model design and selection stages. Motivated by the finding, unlike the previous studies designed for general time-series data, we propose a novel data imputation model to represent patterns and their cyclic rotations in electric load data. Simulation studies reveal that the proposed model works well when the time window size is a divisor of the cycle length, which significantly reduces model selection efforts. Numerical results with two real data sets justify our findings and the performance of the proposed approach against benchmark methods.

Hybrid deep learning of social media big data for predicting the evolution of COVID-19 transmission

In this study, a hybrid deep-learning model termed as ODANN, built upon neural networks (NN) coupled with data assimilation and natural language processing (NLP) features extraction methods, has been constructed to concurrently process daily COVID-19 time-series records and large volumes of COVID-19 related Twitter data, as representative of the global community’s aggregated emotional responses towards the current pandemic, to model the growth rate in the number of confirmed COVID-19 cases globally via a proposed G parameter. Overall, there were 3 key components to ODANN’s development phase, namely: (i) data hydration and pre-processing were performed on COVID-19 related Twitter data ranging between 23 January 2020 and 10 May 2020, which amounted to over 100 million Tweets written in English language; (ii) multiple NLP features extraction methods were subsequently leveraged to encode the hydrated Twitter data into useful semantic word vectors for training ODANN under an optimal set of hyperparameters; and (iii) historical time-series data of defined characteristics were also assimilated into ODANN’s selected hidden layer(s) to model the G parameter daily with a lead-time of 1 day. By far, our experimental results demonstrated that by adopting a rolling time-window size of 5 days, with respect to the number of historical time-series records for assimilating different data features, enabled ODANN to outperform other traditional time-series models and recent studies, in terms of the computed RMSE and MAE scores attained from the model’s testing step. Overall, the summarized results from ODANN demonstrated its competitive edge in modelling and forecasting the growth rate in the number of COVID-19 cases globally.

Influence of Sliding Time Window Size Selection Based on Heart Rate Variability Signal Analysis on Intelligent Monitoring of Noxious Stimulation under Anesthesia.

In recent decades, little progress of objective evaluation of pain and noxious stimulation has been achieved under anesthesia. Some researches based on medical signals have failed to provide a general understanding of this problem. This paper presents a feature extraction method for heart rate variability signals, aiming at further improving the evaluation of noxious stimulation. In the process of data processing, the empirical mode decomposition is used to decompose and recombine heart rate variability signals, and the sliding time window approach is used to extract the signal features of noxious stimulation, respectively. The influence of window size on feature extraction is studied by changing the window size. By comparing the results, the feature extraction in the process of data processing is valuable, and the selection of window size has a significant impact. With the increase of selected window sizes, we can get better detection results. But for the best choice of window size, to ensure the accuracy of the results and to make it easy to use, then, we need to get just a suitable window size.