Time Sequence Data Research Articles

Time-series well performance prediction based on Long Short-Term Memory (LSTM) neural network model

Oil production forecasting is one of the most critical issues during the exploitation phase of the oilfield. The limitations of traditional approaches make time-series production prediction still challenging. With the development of artificial intelligence, high-performance algorithms make reliable production prediction possible from the data perspective. This study proposes a Long Short-Term Memory (LSTM) neural network based model to infer the production of fractured horizontal wells in a volcanic reservoir, which addresses the limitations of traditional method and shows accurate predictions. The LSTM neural network enables to capture the dependencies of the oil rate time sequence data and incorporate the production constraints. Particle Swarm Optimization algorithm (PSO) is employed to optimize the essential configuration of the LSTM model. For evaluation purpose, two case studies are carried out with the production dynamics from synthetic model and from the Xinjiang oilfield, China. Towards a fair evaluation, the performance of the proposed approach is compared with traditional neural networks, time-series forecasting approaches, and conventional decline curves. The experiment results show that the proposed LSTM model outperforms other approaches.

An intelligent healthcare system with data priority based on multi vital biosignals

Background and Objective: Home-based personal healthcare systems are becoming popular and affordable due to the development of Internet of Things (IoT) devices. However, with an increasing number of users, such healthcare systems are challenged to store and process enormous volumes of data. For instance, multi-biosignal data are collected continuously from patients using IoT device like body sensors and are sent to the server by portable devices for further analysis (e.g., knowledge discovery or the clinical event prediction). These enormous amount of data from large number of patients are causing the transmission overhead and high latency in the network which are responsible for inefficiency issues in clinical event prediction. To address these problems, in this paper, data assessment method is introduced to improve the efficiency in data collection and data prediction.Methods: The assessment algorithm is inspired by National Early Warning Score (NEWS) used in Emergency Department. In our method, only the abnormal time-sequence data for analysis are sent to the server. Thus, the waiting time of data before prediction can be optimized because data with higher priority are processed in front of those with lower priority, which helps our system to provide diagnostic decisions in a proper time according to patients’ urgency.Results: Our experiments show that the proposed model ideally can save 20% volume of data in the collection and can reduce 75% waiting time of data with the highest priority before predicting. In addition, the waiting time of data for further analysis is optimized compared to the normal processing flow.Conclusion: The paper introduces an enhanced healthcare system with assessing data priority in order to optimize the data collection and the prediction in terms of data size and waiting time.

Investigating temperature-induced structural changes of lead halide perovskites by in situ X-ray powder diffraction

In situ X-ray diffraction experiments offer a unique opportunity to investigate structural dynamics at atomic resolution, by collecting several patterns in an appropriate time sequence (data matrix) while varying the applied stimulus (e.g. temperature changes). Individual measurements can be processed independently by refinement procedures that are based on prior knowledge of the average structure of each crystal phase present in the sample. If the refinement converges, parameters of the average structural model can be assessed and studied as a function of the stimulus variations. An alternative approach consists in applying a multivariate analysis to the data matrix as a whole. Methods such as principal component analysis (PCA) and phase-sensitive detection perform fast, blind and model-independent calculations that can be used for on-site analysis to identify trends in data actually related to the applied stimulus. Both classical and multivariate approaches are here applied to the in situ X-ray diffraction pair distribution function (PDF) setup on two samples of the hybrid perovskite methylammonium (MA) lead iodide obtained by different synthetic routes, subjected to temperature variations. The PDF refinement allows assessing the occurrence of temperature-induced rotations of the PbI6 octahedra and variations in the relative amount of MAPbI3 and intermediate PbI2–MAI–DMSO (dimethyl sulfoxide) crystal phases. A change in the orientation of the methylammonium molecule with temperature is also characterized. Results of the multivariate analysis tools, which include a newly introduced space-dependent variant of PCA, are described, interpreted and validated against simulated data, and their specificity and relation to refinement results are highlighted. The interaction between nearby octahedra is identified as the driving force for the tetragonal-to-cubic phase transition, and three fundamental trends in data having different temperature behaviours are unveiled: (i) irreversible weight-fraction variations of the MAPbI3 and PbI2–MAI–DMSO phases; (ii) reversible structural changes related to the MAPbI3 crystalline phase and its lattice distortion in the ab plane, having the same frequency as the temperature variations; (iii) reversible lattice distortion along the c axis, occurring at twice the frequency of the temperature changes.

Time Series Prediction Model of Grey Wolf Optimized Echo State Network

As a novel recursion neural network, Echo State Networks (ESN) are characterized by strong nonlinear prediction capability and effective and straightforward training algorithms. However, conventional ESN predictions require a large volume of training samples. Meanwhile, the time sequence data are complicated and unstable, resulting in insufficient learning of this network and difficult training. As a result, the accuracies of conventional ESN predictions are limited. Aimed at this issue, a time series prediction model of Grey Wolf optimized ESN has been proposed. Wout of ESN was optimized using the Grey Wolf algorithm and predictions of time series data were achieved using simplified training. The results indicated that the optimized time series prediction method exhibits superior prediction accuracy at a small sample size, compared with conventional prediction methods.

Ensemble multi‐objective evolutionary algorithm for gene regulatory network reconstruction based on fuzzy cognitive maps

Many methods aim to use data, especially data about gene expression based on high throughput genomic methods, to identify complicated regulatory relationships between genes. The authors employ a simple but powerful tool, called fuzzy cognitive maps (FCMs), to accurately reconstruct gene regulatory networks (GRNs). Many automated methods have been carried out for training FCMs from data. These methods focus on simulating the observed time sequence data, but neglect the optimisation of network structure. In fact, the FCM learning problem is multi-objective which contains network structure information, thus, the authors propose a new algorithm combining ensemble strategy and multi-objective evolutionary algorithm (MOEA), called EMOEA FCM -GRN, to reconstruct GRNs based on FCMs. In EMOEA FCM -GRN, the MOEA first learns a series of networks with different structures by analysing historical data simultaneously, which is helpful in finding the target network with distinct optimal local information. Then, the networks which receive small simulation error on the training set are selected from the Pareto front and an efficient ensemble strategy is provided to combine these selected networks to the final network. The experiments on the DREAM4 challenge and synthetic FCMs illustrate that EMOEA FCM -GRN is efficient and able to reconstruct GRNs accurately.

Smart Fetal Care

The exam of fetal well-being during routine prenatal care plays a crucial role in preventing pregnancy complications and reducing the risks of miscarriages, birth defects and other health problems. However, the conventional prenatal screening and diagnosis is conducted by medical professionals in a clinical environment, which is subject to certain limitations such as manpower, medical devices and location, time and cost of services, etc. This paper presents a new approach to detect and monitor fetal movement safely and reliably without any constrains of time, environment and cost. Unlike the conventional method, our contribution includes a novel soft sensor pad which can automatically detect fetal movement and uterine contraction nonintrusively and the robust data analysis software to monitor pregnancy health and screen abnormalities with quantitative assessment. The monitoring belt embedded with the soft sensor pad is wearable, non-intrusive, radiation free and washable. The new algorithms are robust for noise removal, feature extraction, time sequence data analysis and decision support to achieve personalized care. Both the design of soft sensor pad and functions of the belt are original and unique. The results of preliminary clinical trials demonstrate the feasibility and advantages of our prototype.

河南省产业结构演进和经济增长关系的实证分析——以1978~2017年时间序列数据为例

河南省产业结构演进和经济增长关系的实证分析——以1978~2017年时间序列数据为例

A Simple Recurrent Unit Model Based Intrusion Detection System With DCGAN

Due to the complex and time-varying network environments, traditional methods are difficult to extract accurate features of intrusion behavior from the high-dimensional data samples and process the high-volume of these data efficiently. Even worse, the network intrusion samples are submerged into a large number of normal data packets, which leads to insufficient samples for model training; therefore it is accompanied by high false detection rates. To address the challenge of unbalanced positive and negative learning samples, we propose using deep convolutional generative adversarial networks (DCGAN), which allows features to be extracted directly from the rawdata, and then generates new training-sets by learning from the rawdata. Given the fact that the attack samples are usually intra-dependent time sequence data, we apply long short-term memory (LSTM) to automatically learn the features of network intrusion behaviors. However, it is hard to parallelize the learning/training of the LSTM network, since the LSTM algorithm depends on the result of the previous moment. To remove such dependency and enable intrusion detection in real time, we propose a simple recurrent unit based (SRU)-based model. The proposed model was verified by extensive experiments on the benchmark datasets KDD'99 and NSL-KDD, which effectively identifies normal and abnormal network activities. It achieves 99.73% accuracy on the KDD'99 dataset and 99.62% on the NSL-KDD dataset.

Learning low-complexity autoregressive models via proximal alternating minimization

We consider the estimation of the state transition matrix in vector autoregressive models, when time sequence data is limited but nonsequence steady-state data is abundant. To leverage both sources of data, we formulate the least squares minimization problem regularized by a Lyapunov penalty. We impose cardinality or rank constraints to reduce the complexity of the autoregressive model. The resulting nonconvex, nonsmooth problem is solved by using the proximal alternating linearization method (PALM). We prove that PALM is globally convergent to a critical point and that the estimation error monotonically decreases. Explicit formulas are obtained for the proximal operators to facilitate the implementation of PALM. We demonstrate the effectiveness of the developed method by numerical experiments.

Daily urban water supply modeling and forecasting based on the elliptic orbit model with weekly quasi-periodic extension

ABSTRACTModeling and forecasting for various time horizons of urban water supply are important for different operations within a utility company. This study proposes the ‘elliptic orbit model’ for daily urban water supply prediction from the viewpoint of time-series analysis. As additional efforts and costs are required to acquire and predict more different forecast variables, it is argued that many studies failed to carefully check whether such efforts and costs were deserved and to what degree they might ameliorate the prediction accuracy. Thus predictive modeling based on available water supply data has its own advantages. Only the water-supply time-sequence data is used and mapped into the polar coordinates to design the proposed ‘elliptic orbit model’, so the purpose of this study is to present one vivid approach for forecasting daily urban water supply in an intuitive and concise way. The proposed model archives satisfying experimental results and may inspire other applications.

Chaos sequence data classification mining research under network visualization analysis

ABSTRACTNetwork visualization analysis is a method of quantitative analysis of data in network services by using visualization techniques. However, it is difficult to classify the multimodal underlying sequence characteristics of the data by using the current method for the chaos sequence data classification mining; there is a problem of the large error of the classification of chaotic sequence data. In the article, this method is a method of network visualization analysis of chaotic sequence data classification mining based on multimodal subspace correlation transmission. This method uses the hidden Markov model to convert the chaotic sequence data into the likelihood space, the magnitude of the likelihood is identified by the symmetry Kullback–Leibler distance, extracting the underlying characteristics of multiple modal states, and carry out the transmission of the correlation between video data multimodal subspace. The accuracy and the mean square error were used to define the performance of the chaotic sequence data classification. The classification accuracy of this method is about 90%, and the root mean square error is less than 0.03, which greatly improves the classification accuracy. The packet loss rate of literature [Zheng B, Su H, Lin L. Application of unsupervised feature selection in time series data mining. Chin J Sci Instrum. 2014;35(4):834–840] is about 60%, the missing rate of literature [Mengli R. Algorithm of web hot data mining based on structured segmentation. Bull Sci Technol. 2015;4:115–117] is 50%, and the packet loss rate of this method is about 20%, which proves that this method has higher classification accuracy and application value, and is more suitable for other sequence data mining applications. The similarity between the time sequence data is given and reduces the dimension of the original data in order to obtain the coordinates module in the lower dimensional sequence data space, and then complete network visualization analysis to analyze chaos sequence data mining. The experimental simulation shows that the proposed method has a higher classification precision and greatly expands the application domain of the sequential data classification mining.

A novel spectrum enhancement technique for multi-temporal, multi-spectral data using spatial-temporal filtering

Time-sequence remote sensing images are usually captured under varying acquisition conditions due to atmospheric differences, lighting condition, humidity etc. Comparing the spectral values of the well-registered images taken at a different time is a complicated issue due to the non-linear spectrum distortion caused by these effects. Atmospheric correction can eliminate part of the errors, while precisely removing it requires many other in-situ data such as the weather condition and optical aerosol depth. We propose an algorithm that performs spatial-temporal inferences that correct the spectral values through a data-driven approach - we developed a simple 3D spatiotemporal filtering method that uses the time-sequence imagery themselves to homogenize the spectral property of similar objects while being heterogeneous to objects with significant differences. We have performed extensive experiments using medium resolution Landsat dataset and high-resolution Planet imagery, by evaluating the classification results from both classic machine learning (sample selected from the current image) and transfer learning (samples selected from one dataset and applied to the other dataset). The experiments results show that the proposed 3D spatiotemporal filter can improve the accuracy of classification using transfer learning by ≈5%, ≈15%, and ≈2%. We have also demonstrated that the enhanced time-sequence image offers much better change detection outputs using just a simple image differencing method. The improved results in typical remote sensing tasks indicate our proposed method being effective for time-sequence data preprocessing.

RESEARCH AND APPLICATION OF REMOTE SENSING MONITORING METHOD FOR DESERTIFICATION LAND UNDER TIME AND SPACE CONSTRAINTS

Abstract. Serious land desertification and sandified threaten the urban ecological security and the sustainable economic and social development. In recent years, a large number of mobile sand dunes in Horqin sandy land flow into the northwest of Liaoning Province under the monsoon, make local agriculture suffer serious harm. According to the characteristics of desertification land in northwestern Liaoning, based on the First National Geographical Survey data, the Second National Land Survey data and the 1984–2014 Landsat satellite long time sequence data and other multi-source data, we constructed a remote sensing monitoring index system of desertification land in Northwest Liaoning. Through the analysis of space-time-spectral characteristics of desertification land, a method for multi-spectral remote sensing image recognition of desertification land under time-space constraints is proposed. This method was used to identify and extract the distribution and classification of desertification land of Chaoyang City (a typical citie of desertification in northwestern Liaoning) in 2008 and 2014, and monitored the changes and transfers of desertification land from 2008 to 2014. Sandification information was added to the analysis of traditional landscape changes, improved the analysis model of desertification land landscape index, and the characteristics and laws of landscape dynamics and landscape pattern change of desertification land from 2008 to 2014 were analyzed and revealed.

Fault diagnosis of rolling bearings with recurrent neural network-based autoencoders

As the rolling bearings being the key part of rotary machine, its healthy condition is quite important for safety production. Fault diagnosis of rolling bearing has been research focus for the sake of improving the economic efficiency and guaranteeing the operation security. However, the collected signals are mixed with ambient noise during the operation of rotary machine, which brings great challenge to the exact diagnosis results. Using signals collected from multiple sensors can avoid the loss of local information and extract more helpful characteristics. Recurrent Neural Networks (RNN) is a type of artificial neural network which can deal with multiple time sequence data. The capacity of RNN has been proved outstanding for catching time relevance about time sequence data. This paper proposed a novel method for bearing fault diagnosis with RNN in the form of an autoencoder. In this approach, multiple vibration value of the rolling bearings of the next period are predicted from the previous period by means of Gated Recurrent Unit (GRU)-based denoising autoencoder. These GRU-based non-linear predictive denoising autoencoders (GRU-NP-DAEs) are trained with strong generalization ability for each different fault pattern. Then for the given input data, the reconstruction errors between the next period data and the output data generated by different GRU-NP-DAEs are used to detect anomalous conditions and classify fault type. Classic rotating machinery datasets have been employed to testify the effectiveness of the proposed diagnosis method and its preponderance over some state-of-the-art methods. The experiment results indicate that the proposed method achieves satisfactory performance with strong robustness and high classification accuracy.

Abstract 125: ReflexiveVisual Inspection of Cleft Lip Faces - Analysis of Lookzone Focus Over Time

PURPOSE: Humans reflexively inspect faces. Elucidating facial lookzone focus over time of cleft lip faces may offer insight into the early visual processing of facial normality/abnormality. By delineating the time sequence of visual impression formation, surgeons and their patients may pinpoint the most salient facial features so as to better direct prioritization of surgical reconstruction. METHODS: 179 experimental and 179 control facial images were obtained from the senior author’s practice. Experimental images included 41 individuals with repaired cleft lip, and a variety of other facial diagnoses. 720 subjects rated the images for attractiveness. Twenty standardized lookzone regions were mapped onto each facial image. A separate group of 402 subjects observed the images while an infrared eye-tracking camera continuously recorded their eye movements for 6 seconds. R console TraMineR was utilized to analyze the time sequence data. The gender and personal history of observer facial deformity was recorded. OUTCOMES MEASURED:Image attractiveness was rated on a 1–7 Likert scale. Total number of eye fixations within different lookzone regions was recorded continuously over the 6 seconds viewing period. RESULTS: (i) All observers start focusing on the face after 500ms, and on the cleft defect 200ms after facial scanning, but they revert to the control pattern of focusing on the periorbital area after 2400ms. (ii) Male maintain their focus on cleft defect much longer than females (3600ms vs 1800ms) (iii) Observers with family history of facial deformity maintain their focus on cleft defect throughout the 6000ms viewing period, while those without facial history lose focus on the cleft defect after 2000ms. (iv) The attractiveness ratings of the cleft images had no discernable impact on the sequence of reflexive inspection of the cleft images, except for the most attractive cleft images, for which the lip was not focused upon at all (but rather the nasal deformity). (v) Laterality of the cleft deformity did not impact the sequence of facial inspection, but bilateral clefts were less of a visual draw, with reversion to a control pattern of inspection early, versus a more continued focus on the cleft defect for unilateral clefts. CONCLUSION: Observers are reflexively drawn to the abnormal region of cleft faces upon immediate exposure, before reverting to a more natural pattern of facial inspection after about 2.5 seconds. Unilateral clefts - and cleft faces that are considered less attractive overall - induce a more sustained fixation within the perioral region. A personal history of facial deformity in general appeared to heighten sensitivity for cleft deformity. Males focused longer on the cleft defect compared to females. T. Boonipat: None. S. Prakkamakul: None. K. Fleming: None. M. Stotland: None.

Analysis of medications change in Parkinson’s disease progression data

Parkinson's disease is a neurodegenerative disorder that affects people worldwide. Careful management of patient's condition is crucial to ensure the patient's independence and quality of life. This is achieved by personalized treatment based on individual patient's symptoms and medical history. The aim of this study is to determine patient groups with similar disease progression patterns coupled with patterns of medications change that lead to the improvement or decline of patients' quality of life symptoms. To this end, this paper proposes a new methodology for clustering of short time series of patients' symptoms and prescribed medications data, and time sequence data analysis using skip-grams to monitor disease progression. The results demonstrate that motor and autonomic symptoms are the most informative for evaluating the quality of life of Parkinson's disease patients. We show that Parkinson's disease patients can be divided into clusters ordered in accordance with the severity of their symptoms. By following the evolution of symptoms for each patient separately, we were able to determine patterns of medications change which can lead to the improvement or worsening of the patients' quality of life.

Multi-Spectral Imaging from an Unmanned Aerial Vehicle Enables the Assessment of Seasonal Leaf Area Dynamics of Sorghum Breeding Lines.

Genetic improvement in sorghum breeding programs requires the assessment of adaptation traits in small-plot breeding trials across multiple environments. Many of these phenotypic assessments are made by manual measurement or visual scoring, both of which are time consuming and expensive. This limits trial size and the potential for genetic gain. In addition, these methods are typically restricted to point estimates of particular traits, such as leaf senescence or flowering and do not capture the dynamic nature of crop growth. In water-limited environments in particular, information on leaf area development over time would provide valuable insight into water use and adaptation to water scarcity during specific phenological stages of crop development. Current methods to estimate plant leaf area index (LAI) involve destructive sampling and are not practical in breeding. Unmanned aerial vehicles (UAV) and proximal-sensing technologies open new opportunities to assess these traits multiple times in large small-plot trials. We analyzed vegetation-specific crop indices obtained from a narrowband multi-spectral camera on board a UAV platform flown over a small pilot trial with 30 plots (10 genotypes randomized within 3 blocks). Due to variable emergence we were able to assess the utility of these vegetation indices to estimate canopy cover and LAI over a large range of plant densities. We found good correlations between the Normalized Difference Vegetation Index (NDVI) and the Enhanced Vegetation Index (EVI) with plant number per plot, canopy cover and LAI both during the vegetative growth phase (pre-anthesis) and at maximum canopy cover shortly after anthesis. We also analyzed the utility of time-sequence data to assess the senescence pattern of sorghum genotypes known as fast (senescent) or slow senescing (stay-green) types. The Normalized Difference Red Edge (NDRE) index which estimates leaf chlorophyll content was most useful in characterizing the leaf area dynamics/senescence patterns of contrasting genotypes. These methods to monitor dynamics of green and senesced leaf area are suitable for out-scaling to enhance phenotyping of additional crop canopy characteristics and likely crop yield responses among genotypes across large fields and multiple dates.

ESPRIT-Forest: Parallel clustering of massive amplicon sequence data in subquadratic time.

The rapid development of sequencing technology has led to an explosive accumulation of genomic sequence data. Clustering is often the first step to perform in sequence analysis, and hierarchical clustering is one of the most commonly used approaches for this purpose. However, it is currently computationally expensive to perform hierarchical clustering of extremely large sequence datasets due to its quadratic time and space complexities. In this paper we developed a new algorithm called ESPRIT-Forest for parallel hierarchical clustering of sequences. The algorithm achieves subquadratic time and space complexity and maintains a high clustering accuracy comparable to the standard method. The basic idea is to organize sequences into a pseudo-metric based partitioning tree for sub-linear time searching of nearest neighbors, and then use a new multiple-pair merging criterion to construct clusters in parallel using multiple threads. The new algorithm was tested on the human microbiome project (HMP) dataset, currently one of the largest published microbial 16S rRNA sequence dataset. Our experiment demonstrated that with the power of parallel computing it is now compu- tationally feasible to perform hierarchical clustering analysis of tens of millions of sequences. The software is available at http://www.acsu.buffalo.edu/∼yijunsun/lab/ESPRIT-Forest.html.

Forecasting sea level changes applying data mining techniques to the Cristobal Bay time series, Panama

Time series forecasting using data mining models applied to various time sequence data of a wide variety of domains has been well documented. In this work, time series of water level data recorded every hour at ‘Cristobal Bay’ in Panama during the years 1909–1980 are employed to construct a model(s) that can be suitable for predicting changes in sea level patterns. Four time lag assemblages of variable combinations of the time series information are fully explored to identify the optimal combinations for the dataset using a data mining tool. The results, based on the assessment using time series of Cristobal data, show that in general using cross-validation and a longer time lag period of the time series led to more accurate forecasting of the model than that of a shorter lag period of the time series. The study also suggests that data mining techniques using cross-validation and the aid of an attribute evaluator can be effectively used in modeling time series for changes in sea level at coastal areas, and changes in ecosystems that by their nature are characterized by nonlinearity and presentation of chaotic climatic changes in their physical behavior.

WSN Missing Data Imputing Based on Multiple Time Granularity

Missing data is a common phenomenon in the data collection process of wireless sensor network (WSN), and the missing data imputing is an important issue of WSN stream data mining. Currently WSN missing data imputing method has little considered about the dynamic characteristics of internal data time structure during the data collection process, which makes data imputing difficult to reflect the real monitoring change objectively. In order to analyze the internal structure and dynamics of WSN time sequence data systematically, with the equivalence relation of the monitored object the time domain can be regarded as a series of integral time granule (ie atomic time point set), a wireless sensor network timing information system (WTIS) is established. The system can reason logically at different time granularity, and a multiple optimal time granularity strategy of WTIS based on hierarchical successive approximation approach is proposed. Finally, based on the research, a multiple optimal time granularity WSN missing data clustering imputing algorithm is proposed. Compared with traditional fixed time granularity missing data imputing algorithm, experiments show that the algorithm can lower error rate when imputing WSN missing data.