Dynamic Time Series Data Research Articles

Constructing the CityGML ADE for the Multi-Source Data Integration of Urban Flooding

Urban flooding, as one of the most serious natural disasters, has caused considerable personal injury and property damage throughout the world. To better cope with the problem of waterlogging, the experts have developed many waterlogging models that can accurately simulate the process of pipe network drainage and water accumulation. The study of urban waterlogging involves many data types. These data come from the departments of hydrology, meteorology, planning, surveying, and mapping, etc. The incoordination of space–time scale and format standard has brought huge obstacles to the study of urban waterlogging. This is not conducive to interpretation, transmission, and visualization in today’s network environment. In this paper, the entities and attributes related to waterlogging are defined. Based on the five modules of urban drainage network, sub basin, dynamic water body, time series, and meteorological data, the corresponding UML (Unified Modeling Language) model is designed and constructed. On this basis, the urban waterlogging application domain extension model city waterlogging application domain extension (CTWLADE) is established. According to the characteristics of different types of data, two different methods based on FME object and citygml4j are proposed to realize the corresponding data integration, and KML (Keyhole Markup Language) /glTF data organization form and the corresponding sharing method are proposed to solve the problem that the CTWLADE model data cannot be visualized directly on the web and cannot interact in three-dimensional format. To evaluate the CTWLADE, a prototype system was implemented, which can convert waterlogging-related multi-source data in extensible markup language (XML) files conform. The current CTWLADE can map the data required and provided by the hydraulic software tool storm water management model (SWMM) and is ready to be integrated into a Web 3D Service to provide the data for 3D dynamic visualization in interactive scenes.

Inferring symbolic dynamics of chaotic flows from persistence.

We introduce "state space persistence analysis" for deducing the symbolic dynamics of time series data obtained from high-dimensional chaotic attractors. To this end, we adapt a topological data analysis technique known as persistent homology for the characterization of state space projections of chaotic trajectories and periodic orbits. By comparing the shapes along a chaotic trajectory to those of the periodic orbits, state space persistence analysis quantifies the shape similarity of chaotic trajectory segments and periodic orbits. We demonstrate the method by applying it to the three-dimensional Rössler system and a 30-dimensional discretization of the Kuramoto-Sivashinsky partial differential equation in (1+1) dimensions.

Present and future perspectives of artificial intelligence: examples of mathematical approaches for analysis of disease dynamics

Artificial intelligence (AI) has been applied widely in medicine. For example, deep neural network-based deep learning is particularly effective for pattern recognition in static medical images. Additionally, dynamic time series data are analysed ubiquitously in biology and medicine, as in the application of BCR-ABL International Scale time series data measured from CML patients treated with tyrosine-kinase inhibitors. Nonlinear data analyses, rather than conventional deep learning, can be more powerful for this type of dynamic disease information. Here, I introduce our mathematical approaches that are applicable for disease dynamics, such as dynamical network biomarkers (DNB) and randomly distributed embedding (RDE), as examples of nonlinear data analyses. I also discuss the availability of neuroinspired and neuromorphic hardware systems, which we are developing for potential use in next-generation AI.

A Survey of Time Series Data Visualization Methods

In the era of big data, the general public is more likely to access big data, but they wouldn’t like to analyze the data. Therefore, the traditional data visualization with certain professionalism is not easy to be accepted by the general public living in the fast pace. Under this background, a new general visualization method for dynamic time series data emerges as the times require. Time series data visualization organizes abstract and hard-to-understand data into a form that is easily understood by the public. This method integrates data visualization into short videos, which is more in line with the way people get information in modern fast-paced lifestyles. The modular approach also facilitates public participation in production. This paper summarizes the dynamic visualization methods of time series data ranking, studies the relevant literature, shows its value and existing problems, and gives corresponding suggestions and future research prospects.

Multivariate Analysis on Dynamic Car-Following Data of Non-lane-Based Traffic Environments

The difficulties of the microscopic models in accurate representation of the real traffic phenomena stem from its complexities in the collection and processing of reliable time-series car-following data of non-lane-based traffic environments. Proper estimation of car-following data can suitably ameliorate the realism of traffic sub-models and is still a demanding task. This study describes an image-based in-vehicle trajectory data collection system for the estimation of reliable dynamic time-series data, using camera calibration and in-vehicle GPS information. A copula-based methodological framework is also investigated in this study for evaluating safety in the car-following processes, by accommodating the dependence structure of longitudinal gap, centerline separation and vehicle speeds. Results of the study demonstrated the importance of centerline separation in apprehending the car-following processes. In particular, the probability of maintaining lower gaps increases with the decrease in speed and increase in centerline separation. A 15–20% reduction in the longitudinal gaps is observed for speeds greater than 60 kmph. As importantly, the study recommends the applicability of tri-variate Gaussian copula in assessing the safety or ‘safe distance-keeping’ criteria of drivers in the car-following processes, which can indeed augment the accurate representation of drivers’ behavior and development of the car-following models, advanced driver assistance systems and for safety evaluation in the car-following process of non-lane-based traffic environments.

Data driven nonlinear dynamical systems identification using multi-step CLDNN

In many cases, the equations that dynamical systems are based on are unknown and hard to model and predict. On the other hand, machine learning algorithms are based on the data of a solution as it evolves and do not need equations. In the era of abundant data, using machine learning technology to discover accurate mathematical models of dynamical systems directly from time series data becomes increasingly important. Recently, a multi-step deep neural networks (multi-step DNN) model without need of direct access to temporal gradients is proposed, which can accurately learn the evolution from a given set of observed data, identify nonlinear dynamical systems, and forecast future states. However, the architecture lacks the capability to capture long term temporal dependencies from dynamical time-series data. In the paper, based on the multi-step time-stepping schemes, we proposed a new CLDNN model which combine convolutional layer, long short-term memory layer and fully connected layer, to address the aforementioned weakness. The effectiveness of our model is tested for several benchmark problems involving the identification and prediction of complex, nonlinear and chaotic dynamics. The experiment results show that the multi-step CLDNN has better identification and prediction performance than the multi-step DNN. The research provides possible corroboration for developing new deep learning based algorithms for nonlinear system identification.

Time series non-Gaussian Bayesian bivariate model applied to data on HMPV and RSV: a case of Dadaab in Kenya

BackgroundHuman metapneumovirus (HMPV) have similar symptoms to those caused by the respiratory syncytial virus (RSV). The modes of transmission and dynamics of time series data still remain poorly understood. Climatic factors have long been suspected to be implicated in impacting on the number of cases for these epidemics. Currently, only a few models satisfactorily capture the dynamics of time series data of these two viruses. Our objective was to assess the presence of influence of high incidences between the viruses and to ascertain whether higher incidences of one virus are influenced by the other.MethodsIn this study, we used a negative binomial model to investigate the relationship between RSV and HMPV while adjusting for climatic factors. We specifically aimed at establishing the heterogeneity in the autoregressive effect to account for the influence between these viruses.ResultsIn this study, our findings showed that RSV incidence contributed to the severity of HMPV incidence. This was achieved through comparison of 12 models with different structures, including those with and without interaction between climatic factors. The models with climatic factors out-performed those without.ConclusionsThe study has improved our understanding of the dynamics of RSV and HMPV in relation to climatic cofactors thereby setting a platform to devise better intervention measures to combat the epidemics. We conclude that preventing and controlling RSV infection subsequently reduces the incidence of HMPV.

Multilevel Weighted Feature Fusion Using Convolutional Neural Networks for EEG Motor Imagery Classification

Deep learning methods, such as convolution neural networks (CNNs), have achieved remarkable success in computer vision tasks. Hence, an increasing trend in using deep learning for electroencephalograph (EEG) analysis is evident. Extracting relevant information from CNN features is one of the key reasons behind the success of the CNN-based deep learning models. Some CNN models use convolutional features from different CNN layers with good effect. However, extraction and fusion of multilevel convolutional features remain unexplored for EEG applications. Moreover, cognitive computing and artificial intelligence experience increasing applications in all fields. Cognitive process is based on understanding human brain cognition through signals, such as EEG. Hence, deep learning can aid in developing cognitive systems and related applications by improving EEG decoding. The classification and recognition of EEG have consistently been challenging due to its characteristics of dynamic time series data and low signal-to-noise ratio. However, the information hidden in different convolution layers can aid in improving feature discrimination capability. In this paper, we use the EEG motor imagery data to uncover the benefits of extracting and fusing multilevel convolutional features from different CNN layers, which are abstract representations of the input at various levels. Our proposed CNN model can learn robust spectral and temporal features from the raw EEG data. We demonstrate that such multilevel feature fusion outperforms the models that use features only from the last layer. Our results are better than the state of the art for EEG decoding and classification.

Deriving inferential statistics from recurrence plots: A recurrence-based test of differences between sample distributions and its comparison to the two-sample Kolmogorov-Smirnov test.

Recurrence plots (RPs) have proved to be a very versatile tool to analyze temporal dynamics of time series data. However, it has also been conjectured that RPs can be used to model samples of random variables, that is, data that do not contain any temporal dependencies. In the current paper, we show that RPs can indeed be used to mimic nonparametric inferential statistics. Particularly, we use the case of the two-sample Kolmogorov-Smirnov test as a proof-of-concept, showing how such a test can be done based on RPs. Simulations on differences in mean, variance, and shape of two distributions show that the results of the classical two-sample Kolmogorov-Smirnov test and the recurrence-based test for differences in distributions of two independent samples scale well with each other. While the Kolmogorov-Smirnov test seems to be more sensitive in detecting differences in means, the recurrence based test seems to be more sensitive to detect heteroscedasticity and asymmetry. Potential improvements of our approach as well as extensions to tests with individual distributions are discussed.

Konversi Lahan Sawah di Indonesia dan Pengaruhnya Terhadap Kebijakan Harga Pembelian Pemerintah (HPP) Gabah/Beras

Conversion of paddy fields in Indonesia is quite alarming, reaching an average area of 10 thousand hectares per year. Various regulations have been created and assigned; however, they are considered not to effectively reduce the rate of conversion of the existing paddy fields. The paddy field conversion has brought about negative impacts on the national food security. To overcome these, besides issuing the government regulations to reduce the rate of paddy field conversion, the government has also set a number of policies to boost rice production and to improve the welfare of farmers. One policy is the pricing policy by the Government Purchase Price (grains/rice). This study aimed to analyze the paddy field conversion in Indonesia and its effect on the Government Purchase Price (grains/rice). The data used in this research were the dynamic time series data from the years of 1990-2010, with the consideration of the upward trend in paddy field conversion in Indonesia. The models built an econometric model of simultaneous equations consisted of 41 equations (19 structural equations and 22 identity equations) or in other words, the model consisted of 41 endogenous variables. The results showed that the real prices (grains/rice) of the government purchases have inelastic responses to the changes of all explanatory variables in the short and long terms. The conversion of paddy fields in Indonesia has not only reduced the existing rice production capacity but also has affected the government purchase price policy (grains/rice). The government requires to implement the multi qualities of the government purchasing pries (HPP) for both grains and rice. Keywords: rice, grain, government purchase price, paddy field conversion

Efficient stock price prediction using a Self Evolving Recurrent Neuro-Fuzzy Inference System optimized through a Modified Differential Harmony Search Technique

This paper proposes a new Self Evolving Recurrent Neuro-Fuzzy Inference System (SERNFIS) for efficient prediction of highly fluctuating and irregular financial time series data like stock market indices over varying time frames. The network is modeled including the first order Takagi Sugeno Kang (TSK) type fuzzy if then rules with two types of feedback loops. The recurrent structure in the proposed model comes from locally feeding the firing strength of the fuzzy rule back to itself and by including a few time delay components at the output layer. The novelty of the model is based on the fact that the internal temporal feedback loops and time delayed output feedback loops are used for further enhancing the prediction capability of traditional neuro-fuzzy system in handling more dynamic financial time series data. Another recurrent functional link artificial neural network (RCEFLANN) model is also presented for a comparative study. In the second part of the paper a modified differential harmony search (MDHS) technique is proposed for estimating the parameters of the model including the antecedent, consequent and feedback loop parameters. Experimental results obtained by implementing the model on two different stock market indices demonstrate the effectiveness of the proposed model compared to existing models for stock price prediction.

PhysioSpace: Relating Gene Expression Experiments from Heterogeneous Sources Using Shared Physiological Processes

Relating expression signatures from different sources such as cell lines, in vitro cultures from primary cells and biopsy material is an important task in drug development and translational medicine as well as for tracking of cell fate and disease progression. Especially the comparison of large scale gene expression changes to tissue or cell type specific signatures is of high interest for the tracking of cell fate in (trans-) differentiation experiments and for cancer research, which increasingly focuses on shared processes and the involvement of the microenvironment. These signature relation approaches require robust statistical methods to account for the high biological heterogeneity in clinical data and must cope with small sample sizes in lab experiments and common patterns of co-expression in ubiquitous cellular processes. We describe a novel method, called PhysioSpace, to position dynamics of time series data derived from cellular differentiation and disease progression in a genome-wide expression space. The PhysioSpace is defined by a compendium of publicly available gene expression signatures representing a large set of biological phenotypes. The mapping of gene expression changes onto the PhysioSpace leads to a robust ranking of physiologically relevant signatures, as rigorously evaluated via sample-label permutations. A spherical transformation of the data improves the performance, leading to stable results even in case of small sample sizes. Using PhysioSpace with clinical cancer datasets reveals that such data exhibits large heterogeneity in the number of significant signature associations. This behavior was closely associated with the classification endpoint and cancer type under consideration, indicating shared biological functionalities in disease associated processes. Even though the time series data of cell line differentiation exhibited responses in larger clusters covering several biologically related patterns, top scoring patterns were highly consistent with a priory known biological information and separated from the rest of response patterns.

Online summarization of dynamic time series data

Managing large-scale time series databases has attracted significant attention in the database community recently. Related fundamental problems such as dimensionality reduction, transformation, pattern mining, and similarity search have been studied extensively. Although the time series data are dynamic by nature, as in data streams, current solutions to these fundamental problems have been mostly for the static time series databases. In this paper, we first propose a framework to online summary generation for large-scale and dynamic time series data, such as data streams. Then, we propose online transform-based summarization techniques over data streams that can be updated in constant time and space. We present both the exact and approximate versions of the proposed techniques and provide error bounds for the approximate case. One of our main contributions in this paper is the extensive performance analysis. Our experiments carefully evaluate the quality of the online summaries for point, range, and k–nn queries using real-life dynamic data sets of substantial size.

Analysis of dynamic labeling data

Comprehensive assessments of the organization and regulation of metabolic pathways cannot be limited to steady-state measurements alone but require dynamic time series data. One experimental means of generating such data consists of radioactively labeling precursors and measuring their fate over time. While labeling experiments belong to the standard repertoire of biological laboratory techniques, corresponding mathematical tools for analyzing the non-linear dynamics of tracers are scarce. The article addresses this issue, using Biochemical Systems Theory as the modeling framework. The description of the dynamics of labeled metabolites alone is difficult, but it is demonstrated that these difficulties are easily overcome by setting up dynamic models in two or three blocks, one for the kinetics of the total pools, the second just for the labeled portions, and the third, optional, block for the remaining unlabeled components. Since the dynamic model is not limited in complexity and can account for linear pathways, converging and diverging branches, cycles, and the various observed modes of regulation, the proposed method of non-linear tracer analysis is rather general and permits simulations of most standard labeling experiments, both at steady state and during transients.

Motorway driver behaviour: studies on car following

This paper will report findings of an instrumented vehicle study aimed at assessing one element of driver behaviour, that of car following, on UK motorways. The paper (re-) calibrates one of the most successful of such models—the Action Point model—using dynamic time series data acquired from field tests with an instrumented vehicle. Probability distributions for a number of parameters from the Action Point model are produced and a number of modifications made in order to enhance its value for use in traffic flow and simulation models. Lastly typical headways are compared with existing studies in the area, finding that current headways are far lower than believed. The rationale behind the adoption of such short headways is examined.

A fractional cointegration analysis of the long-run relationship between black and official foreign exchange rates: the case of the Brazilian cruzeiro

This paper applies a relatively new but generalized concept of fractional cointegration to shed some light on the validity of a long-run relationship between monthly black and official US dollar rates of the Brazilian cruzeiro. An investigation of the stochastic properties of these rates reveals that, while the relationship is not cointegrated in their logs, they appear to be fractionally cointegrated if we allow for mean reverting processes that are CI (1, d) with 0< d < 1. The paper demonstrates that relaxing the condition that the residual from the cointegration equation must be a I (0) process, captures a much wider class of mean-reversion behaviour. Furthermore, an analysis of the short-run dynamics propelling the long-run relationship tends to imply that although the official rate influences the changes in the black rate, this is purely over the short-run. In the longer term, the black rate is found to be the initial receptor of any exogenous shock to the equilibrium and it is the official exchange rate that bears the brunt of short-run adjustment to re-establish the long-run equilibrium relationship. The approach illustrated in this paper is shown to hold enormous potential for tests of mean reversion involving hypotheses popular to economic and financial research in general, where the dynamics of time-series data are under constant scrutiny.