Temporal Data Model Research Articles

Precision Marketing Scheme based on Integrating Spatio-temporal Data Clustering and Neural Network

Precision marketing can effectively predict the user’s willingness to consume, with nearly one trillion of market space. The current precision marketing strategy still has remarkable problems such as the low correlation between data model and space; the marketing initial data redundancy and so on. Aiming at the low correlation between the model and space and the initial data redundancy in the marketing, this paper proposes a precise marketing scheme combining the spatiotemporal data clustering and the neural network. Through the establishment of multi - dimensional space - time large data model, effectively enhance the geographic data model, spatio - temporal data model and multi - dimensional data superposition. Design the large-scale data normalization clustering scheme, construct the normalized classification Euclidean distance and the adaptive precision marketing objective function, and realize the K-Means data clustering of the initial data redundancy. Through the neural network based spatial and temporal data precision marketing decision, design data input layer; deal with hidden layer and intelligent output layer; provide an effective program for accurate marketing.

Modeling and estimation for self-exciting spatio-temporal models of terrorist activity

Spatio-temporal hierarchical modeling is an extremely attractive way to model the spread of crime or terrorism data over a given region, especially when the observations are counts and must be modeled discretely. The spatio-temporal diffusion is placed, as a matter of convenience, in the process model allowing for straightforward estimation of the diffusion parameters through Bayesian techniques. However, this method of modeling does not allow for the existence of self-excitation, or a temporal data model dependency, that has been shown to exist in criminal and terrorism data. In this manuscript we will use existing theories on how violence spreads to create models that allow for both spatio-temporal diffusion in the process model as well as temporal diffusion, or self-excitation, in the data model. We will further demonstrate how Laplace approximations similar to their use in Integrated Nested Laplace Approximation can be used to quickly and accurately conduct inference of self-exciting spatio-temporal models allowing practitioners a new way of fitting and comparing multiple process models. We will illustrate this approach by fitting a self-exciting spatio-temporal model to terrorism data in Iraq and demonstrate how choice of process model leads to differing conclusions on the existence of self-excitation in the data and differing conclusions on how violence spread spatially-temporally in that country from 2003–2010.

An Efficient Approach for Modeling Temporal Database with Interval-Based Timestamping in Conventional Database Management Systems

Attaching Date and Time to varying data plays a definite role in representing a dynamic domain and resources on the database systems. The conventional database stores current data and can only represent the knowledge in static sense. Time-varying data representations in relational framework are well described in literature. In this paper we propose an approach for modeling First Normal Form (1NF) and interval-based temporal database over conventional Database Management Systems (DBMS). We choose 1NF approach with interval-based temporal data model for the easily implementation in relational framework as well as to provide the proposed temporal data representation with the modeling and querying power of 1NF relational data model.

An ODT-based abstraction for mining closed sequential temporal patterns in IoT-cloud smart homes

Due to the large amount of usage data collected from smart home appliances in an IoT-cloud environment, efficient mining techniques are of great need to capture the behavioral patterns. Existing mining algorithms are time-consuming and error prone as the amount of data is increasing rapidly. In this paper, we propose an abstraction approach to model temporal data based on an ordered decision tree (ODT) and spatiotemporal characteristics of usage data for IoT-cloud paradigm. The contribution of this research is to provide an efficient representation in terms of average length of patterns, while preserving the spatiotemporal characteristics of original data. We performed extensive experiments on synthetic data to report the performance and provide a comparison with state-of-the-art algorithms to prove the correctness of the proposed technique, even at a low-level of abstraction. The results indicate that the proposed methodology outperform existing techniques due to the inherited power of the ODT temporal structure.

On joint estimation of Gaussian graphical models for spatial and temporal data.

In this article, we first propose a Bayesian neighborhood selection method to estimate Gaussian Graphical Models (GGMs). We show the graph selection consistency of this method in the sense that the posterior probability of the true model converges to one. When there are multiple groups of data available, instead of estimating the networks independently for each group, joint estimation of the networks may utilize the shared information among groups and lead to improved estimation for each individual network. Our method is extended to jointly estimate GGMs in multiple groups of data with complex structures, including spatial data, temporal data, and data with both spatial and temporal structures. Markov random field (MRF) models are used to efficiently incorporate the complex data structures. We develop and implement an efficient algorithm for statistical inference that enables parallel computing. Simulation studies suggest that our approach achieves better accuracy in network estimation compared with methods not incorporating spatial and temporal dependencies when there are shared structures among the networks, and that it performs comparably well otherwise. Finally, we illustrate our method using the human brain gene expression microarray dataset, where the expression levels of genes are measured in different brain regions across multiple time periods.

Ferroan Dolomitization by Seawater Interaction with Mafic Igneous Dikes and Carbonate Host Rock at the Latemar Platform, Dolomites, Italy: Numerical Modeling of Spatial, Temporal, and Temperature Data

Numerous publications address the petrogenesis of the partially dolomitized Latemar carbonate platform, Italy. A common factor is interpretation of geochemical data in terms of heating via regional igneous activity that provided kinetically favorable conditions for replacement dolomitization. New field, petrographic, XRD, and geochemical data demonstrate a spatial, temporal, and geochemical link between replacement dolomite and local mafic igneous dikes that pervasively intrude the platform. Dikes are dominated by strongly altered plagioclase and clinopyroxene. Significantly, where ferroan dolomite is present, it borders dikes. We hypothesize that seawater interacted with mafic minerals, causing Fe enrichment in the fluid that subsequently participated in dolomitization. This hypothesis was tested numerically through thermodynamic (MELTS, Arxim-GEM) and reactive flow (Arxim-LMA) simulations. Results confirm that seawater becomes Fe-enriched during interaction with clinopyroxene (diopside-hedenbergite) and plagioclase (anorthite-albite-orthoclase) solid solutions. Reaction of modified seawater with limestone causes ferroan and nonferroan replacement dolomitization. Dolomite quantities are strongly influenced by temperature. At 40 to 80°C, ferroan dolomite proportions decrease with increasing temperature, indicating that Latemar dolomitization likely occurred at lower temperatures. This relationship between igneous dikes and dolomitization may have general significance due to the widespread association of carbonates with rifting-related igneous environments.

Temporal Database Models Validation and Verification using Mapping Methodology

The majority of any temporal data model is claimed to be superset of the conventional relational data model as well as it has the ability to be represented in the main temporal data models in literature. In this paper, a complete and formal proof is provided in order to show that the A Tuple Timestamping Historical Relational (TTHR) temporal data model is a consistent extension of conventional relational model, and has the expressive power to be represented in other reference temporal data model and the modeling approaches in relational temporal data model.

Learning from heterogeneous temporal data in electronic health records

Electronic health records contain large amounts of longitudinal data that are valuable for biomedical informatics research. The application of machine learning is a promising alternative to manual analysis of such data. However, the complex structure of the data, which includes clinical events that are unevenly distributed over time, poses a challenge for standard learning algorithms. Some approaches to modeling temporal data rely on extracting single values from time series; however, this leads to the loss of potentially valuable sequential information. How to better account for the temporality of clinical data, hence, remains an important research question. In this study, novel representations of temporal data in electronic health records are explored. These representations retain the sequential information, and are directly compatible with standard machine learning algorithms. The explored methods are based on symbolic sequence representations of time series data, which are utilized in a number of different ways. An empirical investigation, using 19 datasets comprising clinical measurements observed over time from a real database of electronic health records, shows that using a distance measure to random subsequences leads to substantial improvements in predictive performance compared to using the original sequences or clustering the sequences. Evidence is moreover provided on the quality of the symbolic sequence representation by comparing it to sequences that are generated using domain knowledge by clinical experts. The proposed method creates representations that better account for the temporality of clinical events, which is often key to prediction tasks in the biomedical domain.

Extending the Kernel of a Relational DBMS with Comprehensive Support for Sequenced Temporal Queries

Many databases contain temporal, or time-referenced, data and use intervals to capture the temporal aspect. While SQL-based database management systems (DBMSs) are capable of supporting the management of interval data, the support they offer can be improved considerably. A range of proposed temporal data models and query languages offer ample evidence to this effect. Natural queries that are very difficult to formulate in SQL are easy to formulate in these temporal query languages. The increased focus on analytics over historical data where queries are generally more complex exacerbates the difficulties and thus the potential benefits of a temporal query language. Commercial DBMSs have recently started to offer limited temporal functionality in a step-by-step manner, focusing on the representation of intervals and neglecting the implementation of the query evaluation engine. This article demonstrates how it is possible to extend the relational database engine to achieve a full-fledged, industrial-strength implementation of sequenced temporal queries, which intuitively are queries that are evaluated at each time point. Our approach reduces temporal queries to nontemporal queries over data with adjusted intervals, and it leaves the processing of nontemporal queries unaffected. Specifically, the approach hinges on three concepts: interval adjustment , timestamp propagation , and attribute scaling . Interval adjustment is enabled by introducing two new relational operators, a temporal normalizer and a temporal aligner, and the latter two concepts are enabled by the replication of timestamp attributes and the use of so-called scaling functions. By providing a set of reduction rules, we can transform any temporal query, expressed in terms of temporal relational operators, to a query expressed in terms of relational operators and the two new operators. We prove that the size of a transformed query is linear in the number of temporal operators in the original query. An integration of the new operators and the transformation rules, along with query optimization rules, into the kernel of PostgreSQL is reported. Empirical studies with the resulting temporal DBMS are covered that offer insights into pertinent design properties of the article's proposal. The new system is available as open-source software.

SYMARMA: a new dynamic model for temporal data on conditional symmetric distribution

Gaussian models of time series, ARMA, have been widely used in the literature. Benjamin et al. (J Am Stat Assoc 98:214–223, 2003) extended these models to the exponential family distributions. Also in that direction, Rocha and Cribari-Neto (Test 18:529–545, 2009) proposed a time series model for the class of beta distributions. In this paper, we develop an autoregressive and moving average symmetric model, named SYMARMA, which is a dynamic model for random variables belonging to the class of symmetric distributions including also a set of regressors. We discuss methods for parameter estimation, hypothesis testing and forecasting. In particular, we provide closed-form expressions for the score function and Fisher information matrix. Robust study is presented based on influence function. We conduct simulation studies to evaluate the consistency and asymptotic normality of the conditional maximum likelihood estimator for the model parameters. An application with real data is presented and discussed.

Gas and heat dynamics of a micro-scaled atmospheric pressure plasma reference jet

Gas and heat dynamics of the ‘Cooperation on Science and Technology (COST) Reference Microplasma Jet’ (COST-jet), a European lead reference device for low temperature atmospheric pressure plasma application, are investigated. Of particular interest to many biomedical application scenarios, the temperature characteristics of a surface impacted by the jet are revealed. Schlieren imaging, thermocouple measurements, infrared thermal imaging and numerical modelling are employed. Temperature spatial profiles in the gas domain reveal heating primarily of the helium fraction of the gas mixture. Thermocouple and model temporal data show a bounded exponential temperature growth described by a single characteristic time parameter to reach ∼63% or (1-1/e) fraction of the temperature increase. Peak temperatures occurred in the gas domain where the carrier jet exits the COST-jet, with values ranging from ambient temperatures to in excess of 100 °C in ‘α-mode’ operation. In a horizontal orientation of the COST-jet a curved trajectory of the helium effluent at low gas flows results from buoyant forces. Gas mixture profiles reveal significant containment of the helium concentrations for a surface placed in close proximity to the COST-jet. Surface heating of a quartz plate follows a similar bounded exponential temporal temperature growth as device heating. Spatial profiles of surface heating are found to correlate strongly to the impacting effluent where peak temperatures occur in regions of maximum surface helium concentration.

The Opportunistic Projection Mining Algorithm in Massive Data

Temporal data model ensures the correctness, validity and compability of data in temporal database, and temporal formal languages model describes the semantic specification of temporal database, both two are core and basis of temporal database system design and development. The paper has exploited the quo of temporal data models and temporal formal languages models, presented FTDM (Formal Temporal Data Model) and L(FTDM) over FTDM by denotation semantics methods. Referring to thinking of software reuse, this paper proposed the definition of languages reuse. Based on the previous works the authors further made temporal formal languages models family {Li} with its particular categorical properties analyzed in this paper by categorical methods, also demonstrated rigorously inherent relationships between temporal formal languages models in {Li}. Their works provided a category theoretic method with universality and flexible expansion for temporal formal languages models, meantime, proposed solid theoretical foundations for design and development of temporal database system.

Asymptotically efficient parameter estimation in hidden Markov spatio-temporal random fields

Estimation of the parameters of Markov random field models for spatial and temporal data arises in many applications. There are computational and sta- tistical challenges in developing efficient estimators because of the complexity of the joint distribution of the spatio-temporal models, especially when they involve hidden states that also need to be estimated from the observations. We develop composite likelihood estimators that are analytically and computationally tractable, and show that they are asymptotically efficient under some mild correlation decay assumptions.

A Cookbook for Temporal Conceptual Data Modelling with Description Logics

We design temporal description logics (TDLs) suitable for reasoning about temporal conceptual data models and investigate their computational complexity. Our formalisms are based onDL-Litelogics with three types of concept inclusions (ranging from atomic concept inclusions and disjointness to the full Booleans), as well as cardinality constraints and role inclusions. The logics are interpreted over the Cartesian products of object domains and the flow of time (ℤ, <), satisfying the constant domain assumption. Concept and role inclusions of the TBox hold at all moments of time (globally), and data assertions of the ABox hold at specified moments of time. To express temporal constraints of conceptual data models, the languages are equipped with flexible and rigid roles, standard future and past temporal operators on concepts, and operators “always” and “sometime” on roles. The most expressive of our TDLs (which can capture lifespan cardinalities and either qualitative or quantitative evolution constraints) turns out to be undecidable. However, by omitting some of the temporal operators on concepts/roles or by restricting the form of concept inclusions, we construct logics whose complexity ranges between NLogSpaceand PSpace. These positive results are obtained by reduction to various clausal fragments of propositional temporal logic, which opens a way to employ propositional or first-order temporal provers for reasoning about temporal data models.

Echo-State Conditional Restricted Boltzmann Machines

Restricted Boltzmann machines (RBMs) are a powerful generative modeling technique, based on a complex graphical model of hidden (latent) variables. Conditional RBMs (CRBMs) are an extension of RBMs tailored to modeling temporal data. A drawback of CRBMs is their consideration of linear temporal dependencies, which limits their capability to capture complex temporal structure. They also require many variables to model long temporal dependencies, a fact that might provoke overfitting proneness. To resolve these issues, in this paper we propose the echo-state CRBM (ES-CRBM): our model uses an echo-state network reservoir in the context of CRBMs to efficiently capture long and complex temporal dynamics, with much fewer trainable parameters compared to conventional CRBMs. In addition, we introduce an (implicit) mixture of ES-CRBM experts (im-ES-CRBM) to enhance even further the capabilities of our ES-CRBM model. The introduced im-ES-CRBM allows for better modeling temporal observations which might comprise a number of latent or observable subpatterns that alternate in a dynamic fashion. It also allows for performing sequence segmentation using our framework. We apply our methods to sequential data modeling and classification experiments using public datasets. As we show, our approach outperforms both existing RBM-based approaches as well as related state-of-the-art methods, such as conditional random fields.

Equipment Fault Prognosis Based on Temporal Association Rules

Equipment fault prognosis is important for reliability, operational safety, and efficient performance of equipment. Temporal fault data model is built according to the principles of the Apriori traditional association rules algorithm based on the characteristics of fault data. An Improved Apriori algorithm and frequent temporal association rules algorithm are proposed in this study by converting fault data to temporal item sets matrix. Equipment fault trends are predicted by mining the frequent temporal association rules of fault data based on the algorithm, which provides good support for equipment maintenance and management. At last an example is given to prove the feasibility and practical application of proposed algorithms. DOI : http://dx.doi.org/10.11591/telkomnika.v12i3.4563 Full Text: PDF

An Indeterminacy Temporal Data Model Based on Probability

There are many kinds of indeterminacy temporal data in temporal database. Therefore, many researchers have focused on building indeterminacy temporal data models. Unfortunately, established models can’t adequately address the challenges posed by indeterminacy temporal information, and can’t adapt to all sorts of involved applications. In this paper, we propose a temporal data model, named BPTM (Temporal model based on probability), to manage the indeterminacy temporal semantics of indeterminacy data. Firstly, we present our tuple-timestamp method to represent and store these temporal data including determinacy and indeterminacy data. Then we introduce the temporal primitives to process temporal relations needed in BPTM. A new probability method is brought forward to get potential information among these indeterminacy data. At last a query example based on CPR (Computer-based Patient Record) is given to show that our method is effective and feasible. DOI: http://dx.doi.org/10.11591/telkomnika.v11i11.3516

THE TEMPORAL DIMENSION IN A 4D ARCHAEOLOGICAL DATA MODEL: APPLICABILITY OF THE GEOINFORMATION STANDARD

Abstract. In recent years, the interest of many researchers in various domains is triggered to move beyond the traditional border of twodimensionality and explore the possibilities of the third and even the fourth, temporal, dimension. The emerging research interest concerning 3D and 4D and the handling of these additional dimensions can bring many benefits to archaeology as well. A 4D GIS tailored to archaeological data would facilitate better insights and more complex analyses. Its basis must be a conceptual 4D archaeological data model, which pays attention to existing data models and standards. Although in some cases more complex, archaeological data are closely related to geography and geo-information. Since the temporal dimension is a, and possibly the most, substantial element in archaeological research, this paper focusses mainly on this dimension. In this paper, the applicability of the ISO 19108 geo-information standard on temporal information for archaeological data is investigated. For a set of common temporal categories, e.g. the excavation time, the appropriate description according to this standard is determined. This will indicate in which cases the internationally recognized standard is suitable for use in an archaeological data model. Furthermore, three versions of the West European archaeological time scale as temporal ordinal reference system are constructed. For the first version, the ISO 19108 structure is used, whereas the second and third are based on geological variants. The results of the performed analysis are favourable to the usability of the ISO 19108 standard in archaeology; however, other temporal standards or data models may yield up better results.

Wavelet principal component analysis of fetal movement counting data preceding hospital examinations due to decreased fetal movement: a prospective cohort study

BackgroundFetal movement (FM) counting is a simple and widely used method of assessing fetal well-being. However, little is known about what women perceive as decreased fetal movement (DFM) and how maternally perceived DFM is reflected in FM charts.MethodsWe analyzed FM counting data from 148 DFM events occurring in 137 pregnancies. The women counted FM daily from pregnancy week 24 until birth using a modified count-to-ten procedure. Common temporal patterns for the two weeks preceding hospital examination due to DFM were extracted from the FM charts using wavelet principal component analysis; a statistical methodology particularly developed for modeling temporal data with sudden changes, i.e. spikes that are frequently found in FM data. The association of the extracted temporal patterns with fetal complications was assessed by including the individuals’ scores on the wavelet principal components as explanatory variables in multivariable logistic regression analyses for two outcome measures: (i) complications identified during DFM-related consultations (n = 148) and (ii) fetal compromise at the time of consultation (including relevant information about birth outcome and placental pathology). The latter outcome variable was restricted to the DFM events occurring within 21 days before birth (n = 76).ResultsAnalyzing the 148 and 76 DFM events, the first three main temporal FM counting patterns explained 87.2% and 87.4%, respectively, of all temporal variation in the FM charts. These three temporal patterns represented overall counting times, sudden spikes around the time of DFM events, and an inverted U-shaped pattern, explaining 75.3%, 8.6%, and 3.3% and 72.5%, 9.6%, and 5.3% of variation in the total cohort and subsample, respectively. Neither of the temporal patterns was significantly associated with the two outcome measures.ConclusionsAcknowledging that sudden, large changes in fetal activity may be underreported in FM charts, our study showed that the temporal FM counting patterns in the two weeks preceding DFM-related consultation contributed little to identify clinically important changes in perceived FM. It thus provides insufficient information for giving detailed advice to women on when to contact health care providers. The importance of qualitative features of maternally perceived DFM should be further explored.

Comparisons of three data storage models in parametric temporal databases

The parametric temporal data model captures a real world entity in a single tuple, which reduces query language complexity. Such a data model, however, is difficult to be implemented on top of conventional databases because of its unfixed attribute sizes. XML is a matured technology and can be an elegant solution for such challenge. Representing data in XML trigger a question about storage efficiency. The goal of this work is to provide a straightforward answer to such a question. To this end, we compare three different storage models for the parametric temporal data model and show that XML is not worse than any other approaches. Furthermore, XML outperforms the other storages under certain conditions. Therefore, our simulation results provide a positive indication that the myth about XML is not true in the parametric temporal data model.