scholarly journals Multi-version Tensor Completion for Time-delayed Spatio-temporal Data

2021 ◽  
Author(s):  
Cheng Qian ◽  
Nikos Kargas ◽  
Cao Xiao ◽  
Lucas Glass ◽  
Nicholas Sidiropoulos ◽  
...  
Keyword(s):  
Real World ◽  
Mean Squared Error ◽  
Ground Truth ◽  
Low Rank ◽  
Tensor Model ◽  
Temporal Data ◽  
Tensor Completion ◽  
Spatio Temporal ◽  
Tensor Data ◽  
Over Time

Real-world spatio-temporal data is often incomplete or inaccurate due to various data loading delays. For example, a location-disease-time tensor of case counts can have multiple delayed updates of recent temporal slices for some locations or diseases. Recovering such missing or noisy (under-reported) elements of the input tensor can be viewed as a generalized tensor completion problem. Existing tensor completion methods usually assume that i) missing elements are randomly distributed and ii) noise for each tensor element is i.i.d. zero-mean. Both assumptions can be violated for spatio-temporal tensor data. We often observe multiple versions of the input tensor with different under-reporting noise levels. The amount of noise can be time- or location-dependent as more updates are progressively introduced to the tensor. We model such dynamic data as a multi-version tensor with an extra tensor mode capturing the data updates. We propose a low-rank tensor model to predict the updates over time. We demonstrate that our method can accurately predict the ground-truth values of many real-world tensors. We obtain up to 27.2% lower root mean-squared-error compared to the best baseline method. Finally, we extend our method to track the tensor data over time, leading to significant computational savings.

scholarly journals Outlier-Robust Multi-Aspect Streaming Tensor Completion and Factorization

2019 ◽  
Author(s):  
Mehrnaz Najafi ◽  
Lifang He ◽  
Philip S. Yu
Keyword(s):  
Real World ◽  
Missing Values ◽  
Real Data ◽  
Low Rank ◽  
Tensor Factorization ◽  
Tensor Completion ◽  
Smooth Optimization ◽  
Novel Method ◽  
Real World Datasets ◽  
Tensor Data

With the increasing popularity of streaming tensor data such as videos and audios, tensor factorization and completion have attracted much attention recently in this area. Existing work usually assume that streaming tensors only grow in one mode. However, in many real-world scenarios, tensors may grow in multiple modes (or dimensions), i.e., multi-aspect streaming tensors. Standard streaming methods cannot directly handle this type of data elegantly. Moreover, due to inevitable system errors, data may be contaminated by outliers, which cause significant deviations from real data values and make such research particularly challenging. In this paper, we propose a novel method for Outlier-Robust Multi-Aspect Streaming Tensor Completion and Factorization (OR-MSTC), which is a technique capable of dealing with missing values and outliers in multi-aspect streaming tensor data. The key idea is to decompose the tensor structure into an underlying low-rank clean tensor and a structured-sparse error (outlier) tensor, along with a weighting tensor to mask missing data. We also develop an efficient algorithm to solve the non-convex and non-smooth optimization problem of OR-MSTC. Experimental results on various real-world datasets show the superiority of the proposed method over the baselines and its robustness against outliers.

scholarly journals 2½-minute 3D 7T 31P-MRSI of the human heart using concentric rings (CRT)

2021 ◽  
Author(s):  
William T Clarke ◽  
Lukas Hingerl ◽  
Bernhard Strasser ◽  
Wolfgang Bogner ◽  
Ladislav Valkovic ◽  
...  
Keyword(s):  
Signal To Noise Ratio ◽  
Variable Number ◽  
Low Rank ◽  
Acquisition Time ◽  
Temporal Data ◽  
Scan Time ◽  
Spread Function ◽  
3D Acquisition ◽  
Spatio Temporal ◽  
31P Mrsi

A 3D density-weighted concentric ring trajectory (CRT) MRSI sequence is implemented for cardiac 31P-MRS at 7T. The point-by-point k-space sampling of traditional phase-encoded CSI sequences severely restricts the minimum scan time at higher spatial resolutions. Our proposed CRT sequence implements a stack of concentric rings trajectory, with a variable number of rings and planes spaced to optimise the density of k-space weighting. This creates flexibility in acquisition time, allowing acquisitions substantially faster than traditional phase-encoded CSI sequences, while retaining high SNR. We first characterise the signal-to-noise ratio and point spread function of the CRT sequence in phantoms. We then evaluate it at five different acquisition times and spatial resolutions in the hearts of five healthy participants at 7T. These different sequence durations are compared with existing published 3D acquisition-weighted CSI sequences with matched acquisition times and spatial resolutions. To minimise the effect of noise on the short acquisitions, low-rank denoising of the spatio-temporal data was also performed after acquisition. The proposed sequence measures 3D localised PCr/ATP ratios of the human myocardium in 2.5 minutes, 2.6 times faster than the minimum scan time for the acquisition-weighted phase-encoded CSI. Alternatively, in the same scan time a 1.7-times smaller nominal voxel volume can be achieved. Low-rank denoising reduced the variance of measured PCr/ATP ratios by 11% across all protocols. The faster acquisitions permitted by 7T CRT 31P-MRSI could make cardiac stress protocols or creatine kinase rate measurements (which involve repeated scans) more tolerable for patients without sacrificing spatial resolution.

scholarly journals Suitability of Graph Database Technology for the Analysis of Spatio-Temporal Data

2020 ◽  
Vol 12 (5) ◽  
pp. 78 ◽  
Author(s):  
Sedick Baker Effendi ◽  
Brink van der Merwe ◽  
Wolf-Tilo Balke
Keyword(s):  
Data Analysis ◽  
Real World ◽  
Spatial Data ◽  
Response Times ◽  
Graph Database ◽  
Temporal Data ◽  
Extensive Evaluation ◽  
Database Technology ◽  
Spatio Temporal ◽  
Real World Datasets

Every day large quantities of spatio-temporal data are captured, whether by Web-based companies for social data mining or by other industries for a variety of applications ranging from disaster relief to marine data analysis. Making sense of all this data dramatically increases the need for intelligent backend systems to provide realtime query response times while scaling well (in terms of storage and performance) with increasing quantities of structured or semi-structured, multi-dimensional data. Currently, relational database solutions with spatial extensions such as PostGIS, seem to come to their limits. However, the use of graph database technology has been rising in popularity and has been found to handle graph-like spatio-temporal data much more effectively. Motivated by the need to effectively store multi-dimensional, interconnected data, this paper investigates whether or not graph database technology is better suited when compared to the extended relational approach. Three database technologies will be investigated using real world datasets namely: PostgreSQL, JanusGraph, and TigerGraph. The datasets used are the Yelp challenge dataset and an ambulance response simulation dataset, thus combining real world spatial data with realistic simulations offering more control over the dataset. Our extensive evaluation is based on how each database performs under practical data analysis scenarios similar to those found on enterprise level.

scholarly journals Monitoring Agricultural Fields Using Sentinel-1 and Temperature Data in Peru: Case Study of Asparagus (Asparagus officinalis L.)

2020 ◽  
Vol 12 (12) ◽  
pp. 1993
Author(s):  
Cristian Silva-Perez ◽  
Armando Marino ◽  
Iain Cameron
Keyword(s):  
Mean Squared Error ◽  
Ground Truth ◽  
Asparagus Officinalis ◽  
Tropical Region ◽  
Start Date ◽  
Ground Station ◽  
Production Cycles ◽  
Data Source ◽  
Spatio Temporal ◽  
Starting Date

This paper presents the analysis and a methodology for monitoring asparagus crops from remote sensing observations in a tropical region, where the local climatological conditions allow farmers to grow two production cycles per year. We used the freely available dual-polarisation GRD data provided by the Sentinel-1 satellite, temperature from a ground station and ground truth from January to August of 2019 to perform the analysis. We showed how particularly the VH polarisation can be used for monitoring the canopy formation, density and the growth rate, revealing connections with temperature. We also present a multi-output machine learning regression algorithm trained on a rich spatio-temporal dataset in which each output estimates the number of asparagus stems that are present in each of the pre-defined crop phenological stages. We tested several scenarios that evaluated the importance of each input data source and feature, with results that showed that the methodology was able to retrieve the number of asparagus stems in each crop stage when using information about starting date and temperature as predictors with coefficients of determination ( R 2 ) between 0.84 and 0.86 and root mean squared error (RMSE) between 2.9 and 2.7. For the multitemporal SAR scenario, results showed a maximum R 2 of 0.87 when using up to 5 images as input and an RMSE that maintains approximately the same values as the number of images increased. This suggests that for the conditions evaluated in this paper, the use of multitemporal SAR data only improved mildly the retrieval when the season start date and accumulated temperature are used to complement the backscatter.

scholarly journals Dimensional Inconsistency Measures and Postulates in Spatio-Temporal Databases

2021 ◽  
Vol 71 ◽  
Author(s):  
John Grant ◽  
Maria Vanina Martinez ◽  
Cristian Molinaro ◽  
Francesco Parisi
Keyword(s):  
Real World ◽  
Knowledge Bases ◽  
Propositional Knowledge ◽  
Location Based Services ◽  
Temporal Databases ◽  
Temporal Data ◽  
Time Space ◽  
Spatio Temporal ◽  
Temporal Dimensions ◽  
Natural Measures

The problem of managing spatio-temporal data arises in many applications, such as location-based services, environmental monitoring, geographic information systems, and many others. Often spatio-temporal data arising from such applications turn out to be inconsistent, i.e., representing an impossible situation in the real world. Though several inconsistency measures have been proposed to quantify in a principled way inconsistency in propositional knowledge bases, little effort has been done so far on inconsistency measures tailored for the spatio-temporal setting. In this paper, we define and investigate new measures that are particularly suitable for dealing with inconsistent spatio-temporal information, because they explicitly take into account the spatial and temporal dimensions, as well as the dimension concerning the identifiers of the monitored objects. Specifically, we first define natural measures that look at individual dimensions (time, space, and objects), and then propose measures based on the notion of a repair. We then analyze their behavior w.r.t. common postulates defined for classical propositional knowledge bases, and find that the latter are not suitable for spatio-temporal databases, in that the proposed inconsistency measures do not often satisfy them. In light of this, we argue that also postulates should explicitly take into account the spatial, temporal, and object dimensions and thus define “dimension-aware” counterparts of common postulates, which are indeed often satisfied by the new inconsistency measures. Finally, we study the complexity of the proposed inconsistency measures.

scholarly journals Analyzing Essential Biodiversity Variables with the VAT System

2019 ◽  
Vol 3 ◽  
Author(s):  
Christian Beilschmidt ◽  
Johannes Drönner ◽  
Néstor Fernández ◽  
Christian Langer ◽  
Michael Mattig ◽  
...  
Keyword(s):  
Processing System ◽  
Easy Access ◽  
Time Range ◽  
Use Case ◽  
Temporal Data ◽  
Area Of Interest ◽  
Essential Biodiversity Variables ◽  
The Creation ◽  
Spatio Temporal ◽  
Over Time

The Essential Biodiversity Variables (EBVs) are important information sources for scientists and decision makers. They are developed and promoted by the Group on Earth Observations Biodiversity Observation Network (GEO BON) together with the community. EBVs provide an abstraction level between measurements and indicators. This enables access to biodiversity observations and allows different groups of users to detect temporal trends as well as regional deviations. In particular, the analysis of EBVs supports finding countermeasures for current important challenges like biodiversity loss and climate change. A visual assessment is an intuitive way to drive the analysis. As one example, researchers can recognize and interpret the changes of forest cover maps over time. The VAT System, in which VAT is an acronym for visualization, analysis and transformation, is an ideal candidate platform for the creation of such an analytical application. It is a geographical processing system that supports a variety of spatio-temporal data types and allows computations using heterogeneous data. For user interaction, it offers a web-based user interface that is built with state-of-the-art web technology. Users can perform interactive analysis of spatio-temporal data by visualizing data on maps and using various graphs and diagrams that are linked to the user’s area of interest. Furthermore, users are enabled to browse through the temporal dimension of the data using a time slider tool. This provides an easy access to large spatio-temporal data sets. One exemplary use case is the creation of EBV statistics for selected countries or areas. This functionality is provided as an app that is built upon the VAT System. Here, users select EBVs, a time range and a metric, and create temporal charts that display developments over time. The charts are constructed internally by employing R scripts that were created by domain experts. The scripts are executed using VAT’s R connectivity module. Finally, users can export the results to their local computers. An export contains the result itself and additionally, a list of citations of the included EBVs as well as a workflow description of all processing steps for reasons of reproducibility. Such a use case exemplifies the suitability of the VAT System to facilitate the creation of similar projects or applications without the need of programming, using VAT’s modular and flexible components.

scholarly journals Μέθοδοι προσπέλασης και επεξεργασίας ερωτήσεων σε χρονικές και χωροχρονικές βάσεις δεδομένων

2002 ◽  
Author(s):  
Θεόδωρος Τζουραμάνης
Keyword(s):  
Real World ◽  
Spatial Data ◽  
Spatial Information ◽  
Temporal Databases ◽  
Regional Data ◽  
Access Methods ◽  
Storage And Retrieval ◽  
Spatio Temporal ◽  
Stored Information ◽  
Over Time

Time is a very important concept related to almost all phenomena of the real world. Information and data correspond to specific time-points and usually change over time. One of the roles of databases is the support of the time evolving nature of the phenomena they model. This ability is of fundamental importance in many applications, such as accounting, banking, law, medical, commercial, econometrics, land and cartographic applications. Temporal and spatio-temporal databases are two categories of databases, which equally deal with the concept of time but are, however, related to different types of applications. Conventional databases have been designed to maintain only the most recently stored information that is current information. As this information is updated, the database content is modified and the last stored information is removed from the database. Therefore, the only retained version of the database is the current one. Temporal databases, on the other hand, support the maintenance of time-evolving data and the satisfaction of specialized queries that are related to three notions of time for these data: the past, the current and the present. Traditional spatial databases are restricted to represent, store and manipulate only static spatial data, such as points, lines, surfaces, volumes and hyper-volumes in multi-dimensional space. However, there are many applications that demand the storage and retrieval of continuously changing spatial information Geographical information systems, image and multi- media databases, urban planning, transportation, mobile communications, computer-aided design and medical databases are only some of the applications that would benefit from the management of this type of dynamically-changing spatial information. Spatio-temporal databases manipulate spatial data, the geometry of which changes dynamically. They provide the chronological framework for the efficient storage and retrieval of all the states of a spatial database over time. This includes the current and past states and the support of spatial queries that refer to present and past time-points as well. In this doctoral dissertation, the research over the temporal and spatio-temporal databases focuses on data that are indexed according to transaction time. More specifically, with regards to spatio-temporal databases, the present research focuses in time-evolving regional data. Real world examples of such applications include the storage and manipulation of data of meteorological phenomena (e.g. atmospheric pressure-zones; icebergs as they change and move over time), of faunal phenomena (e.g. movements of populations of animals/birds/fishes), of urban phenomena (e.g. traffic jams or traffic networks in big cities; city planning events: building and destroying), of natural catastrophes (e.g. fires; hurricanes; oil slicks; floods; pollution clouds) etc. In particular, the focus of the present dissertation is on designing efficient access methods and query processing algorithms for transaction-time databases and databases for time- evolving regional data. This contribution is considered to be of particular importance because access methods play a very important role in the development of efficient database management systems. One access method for transaction-time data and four access methods for time-evolving regional data are designed and implemented. Are also implemented efficient algorithms for the processing of three queries for temporal and five new queries for spatio-temporal databases. These queries exploit the advantage of the properties of these new access methods. The first in the bibliography generator for synthetic time-evolving regional data is also introduced. Finally, an extensive experimental performance evaluation and comparison of all the above four new access methods for time-evolving regional data, is presented. Because of the lack of real benchmark data, the regional data sets used in the experiments were synthetic raster images with real-world semantics that were generated by the new synthetic data generator. The comparison is made under a common and flexible benchmarking environment in order to make it possible to choose the best technique depending on the application and on the characteristics of the manipulated images.

Sign in / Sign up

Export Citation Format

Share Document