Find research datasets worth reusing
Search datasets from major research repositories and use ShareScore to quickly assess how well each record supports discovery, access, and reuse.
30
datasets available to search
ShareScore release 0.9.0
Dataset results
30 results for “Time series prediction”
Time series of in situ Uv-Vis absorbance spectra and high-frequency predictions of total and soluble Fe and Mn concentrations measured at multiple depths in Falling Creek Reservoir (Vinton, VA, USA) in 2020 and 2021
High-frequency measurements of light absorbance were collected at multiple depths in Falling Creek Reservoir (FCR; Vinton, VA, USA) using a s::can Spectrolyser UV-Visible spectrophotometer coupled with a multiplexor pumping system. The system pumps water samples from individual depths into a flow-through cuvette where the UV-vis absorbance spectra of the sample are measured by the spectrophotometer. The system used in our study collected measurements of light absorbance every 2.5 nm wavelengths from 200 nm to 732.5 nm (optical path length of 10 mm) approximately at an hourly time step for seven monitoring depths in the reservoir. Data was collected during two periods; the first deployment (16 October to 9 November 2020) was to observe changes in Fe and Mn concentrations before, during, and after reservoir fall turnover and the second deployment (26 May to 21 June 2021) was to observe the effects of engineered hypolimnetic oxygenation on Fe and Mn concentrations. Partial least squares regression models were developed to generate predictions of total and soluble Fe and Mn concentrations based on the correlation between absorbance spectra and sampling data.
EO4WildFires: An Earth Observation multi-sensor, time-series machine-learning-ready benchmark dataset for wildfire impact prediction
<p>This paper presents a benchmark dataset called EO4WildFires; a multi-sensor (multi spectral; Sentinel-2, Synthetic-Aperture Radar - SAR; Sentinel-1, meteorological parameters; NASA Power) time-series dataset that spans 45 countries, which can be used for developing machine learning and deep learning methods targeted for the estimation of the area that a forest wildfire might cover.</p> <p>This novel EO4WildFires dataset is annotated using EFFIS (European Forest Fire Information System) as forest fire detection and size estimation data source. A total of 31,742 wildfire events are gathered from 2018 to 2022. For each event, Sentinel-2 (multispectral), Sentinel-1 (SAR) and meteorological data are assembled into a single data cube. The meteorological parameters that are included in the data cube are: ratio of actual partial pressure of water vapor to the partial pressure at saturation, average temperature, bias corrected average total precipitation, average wind speed, fraction of land covered by snowfall, percent of root zone soil wetness, snow depth, snow precipitation, as well as percent of soil moisture.</p> <p>The main problem that this dataset is designed to address, is the severity forecasting before wildfires occur. The dataset is not used to predict wildfire events, but rather to predict the severity (size of area damaged by fire) of a wildfire event, if that happens in a specific place under the current and historical forest status, as recorded from multispectral and SAR images, and meteorological data.</p> <p>Using the data cube for the collected wildfire events, the EO4WildFires dataset is used to realize three (3) different preliminary experiments, in order to evaluate the contributing factors for wildfire severity prediction. The first experiment evaluates wildfire size using only the meteorological parameters, the second one utilizes both the multispectral and SAR parts of the dataset, while the third exploits all dataset parts. In each experiment, machine learning models are developed, and their accuracy is evaluated.</p>
A Meta-Learner Approach to Multistep-Ahead Time Series Prediction
<p><strong>Abstract</strong></p> <p>The application of machine learning has become commonplace for problems in modern data science. The democratization of the decision process when choosing a machine learning algorithm has also received considerable attention through the use of meta features and automated machine learning for both classification and regression type problems. However, this is not the case for multistep-ahead time series problems. Time series models generally rely upon the series itself to make future predictions, as opposed to independent features used in regression and classification problems. The structure of a time series is generally described by features such as trend, seasonality, cyclicality, and irregularity. In this research, we demonstrate how time series metrics for these features, in conjunction with an ensemble based regression learner, were used to predict the standardized mean square error of candidate time series prediction models. These experiments used datasets that cover a wide feature space and enable researchers to select the single best performing model or the top N performing models. A robust evaluation was carried out to test the learner's performance on both synthetic and real time series. </p> <p><strong>Proposed Dataset</strong></p> <p>The dataset proposed here gives the results for 20 step ahead predictions for eight Machine Learning/Multi-step ahead prediction strategies for 5,842 time series datasets outlined <a href="https://www.sciencedirect.com/science/article/pii/S2215016121002521">here</a>. It was used as the training data for the Meta Learners in this research. The meta features used are columns C to AE. Columns AH outlines the method/strategy used and columns AI to BB (the error) is the outcome variable for each prediction step. The description of the method/strategies is as follows:</p> <p><strong>Machine Learning methods:</strong></p> <ul> <li>NN: Neural Network</li> <li>ARIMA: Autoregressive Integrated Moving Average</li> <li>SVR: Support Vector Regression</li> <li>LSTM: Long Short Term Memory</li> <li>RNN: Recurrent Neural Network</li> </ul> <p><strong>Multistep ahead prediction strategy:</strong></p> <ul> <li>OSAP: One Step ahead strategy</li> <li>MRFA: Multi Resolution Forecast Aggregation</li> </ul>
Ensemble Machine Learning Prediction of Potential FAPAR: Monthly time-series 2021 and Long-Term Comparison with Actual FAPAR
<p><strong>General Description</strong></p> <p>The dataset contains composites at 250 m spatial resolution of (1) monthly potential FAPAR for the year 2021 from ensemble ML model predictions, (2) the model deviance for each prediction, (3) the yearly average of potential FAPAR, (4) the yearly average of actual FAPAR and (5) the yearly average of the difference between actual and potential (actual minus potential) FAPAR. The dataset is based on the <a href="https://zenodo.org/record/8392976">95th percentile of the monthly aggregated FAPAR</a> derived from <a href="http://glass.umd.edu/Overview.html">250 m 8 d GLASS V6 FAPAR</a>. Potential FAPAR was predicted by fitting an ensemble ML model using globally distributed training points (cca 3 Mio) and a set of 52 biophysical covariates including several layers related to human pressure. The code for modeling potential FAPAR is openly available at <a href="http://github.com/Open-Earth-Monitor/Global_FAPAR_250m">https://github.com/Open-Earth-Monitor/Global_FAPAR_250m</a>. The dataset can be used in many applications like land degradation modeling, land productivity mapping, and land potential mapping. </p> <p><strong>Data Details</strong></p> <ul> <li><strong>Time period:</strong> January 2021 - December 2021</li> <li><strong>Type of data: </strong>Fraction of Absorbed Photosynthetically Active Radiation (FAPAR)</li> <li><strong>How the data was collected or derived:</strong> Derived from 250m 8 d GLASS V6 FAPAR</li> <li><strong>Statistical methods used: </strong>Ensemble machine learning</li> <li><strong>Limitations or exclusions in the data: </strong>The dataset does not include data for Antarctica.</li> <li><strong>Coordinate reference system:</strong> EPSG:4326</li> <li><strong>Bounding box (Xmin, Ymin, Xmax, Ymax):</strong> (-180.00000, -62.0008094, 179.9999424, 87.37000)</li> <li><strong>Spatial resolution:</strong> 1/480 d.d. = 0.00208333 (250m)</li> <li><strong>Image size: </strong>172,800 x 71,698</li> <li><strong>File format: </strong>Cloud Optimized Geotiff (COG) format.</li> </ul> <p><strong>Support</strong></p> <p>If you discover a bug, artifact, or inconsistency, or if you have a question please raise a GitHub issue: <a href="https://github.com/Open-Earth-Monitor/Global_FAPAR_250m/issues">https://github.com/Open-Earth-Monitor/Global_FAPAR_250m/issues</a></p> <p><strong>Reference</strong></p> <p>Hackländer, J., Parente, L., Ho, Y.-F., Hengl, T., Simoes, R., Consoli, D., Şahin, M., Tian, X., Herold, M., Jung, M., Duveiller, G., Weynants, M., Wheeler, I., (2023?) "Land potential assessment and trend-analysis using 2000–2021 FAPAR monthly time-series at 250 m spatial resolution", submitted to PeerJ, preprint available at: <a href="https://doi.org/10.21203/rs.3.rs-3415685/v1">https://doi.org/10.21203/rs.3.rs-3415685/v1</a></p> <p> </p> <p><strong>Name convention</strong></p> <p>To ensure consistency and ease of use across and within the projects, we follow the standard Open-Earth-Monitor file-naming convention. The convention works with 10 fields that describes important properties of the data. In this way users can search files, prepare data analysis etc, without needing to open files. The fields are:</p> <ol> <li><strong>generic variable name:</strong> pot.fapar = Potential Fraction of Absorbed Photosynthetically Active Radiation</li> <li><strong>variable procedure combination: </strong>eml = ensemble machine learning</li> <li><strong>Position in the probability distribution / variable type:</strong> m = mean</li> <li><strong>Spatial support:</strong> 250m</li> <li><strong>Depth reference: </strong>s = surface</li> <li><strong>Time reference begin time:</strong> 20210101 = 2021-01-01</li> <li><strong>Time reference end time:</strong> 20211231 = 2021-12-31</li> <li><strong>Bounding box: </strong>go = global (without Antarctica)</li> <li><strong>EPSG code:</strong> epsg.4326 = EPSG:4326</li> <li><strong>Version code:</strong> v20230924 = 2023-09-24 (creation date)</li> </ol>
WildfireSpreadTS: A dataset of multi-modal time series for wildfire spread prediction
<p>We present a <strong>multi-temporal</strong>, <strong>multi-modal</strong> remote-sensing dataset for predicting <strong>how active wildfires will spread</strong> at a resolution of 24 hours. The dataset consists of <strong>13.607 images</strong> across 607 fire events in the United States from January 2018 to October 2021. For each fire event, the dataset contains a <strong>full time series of daily observations</strong>, containing detected active fires and variables related to <strong>fuel, topography and weather conditions</strong>.</p><h2>Documentation</h2><p><i><strong>WildfireSpreadTS_Documentation.pdf</strong></i> includes further details about the dataset, following Gebru et al.'s <strong>"Datasheets for Datasets"</strong> framework. This documentation is similar to the supplementary material of the associated NeurIPS paper, excluding only information about experimental setup and results. For full details, please refer to the associated paper. </p><h2>Code: Getting started</h2><p>Get started working with the dataset at <a href="https://github.com/SebastianGer/WildfireSpreadTS">https://github.com/SebastianGer/WildfireSpreadTS</a>. </p><p>The code includes a <strong>PyTorch Dataset</strong> and <strong>Lightning DataModule </strong>to allow for easy access. We recommend converting the GeoTIFF files provided here to HDF5 files (bigger files, but much faster). The necessary code is also available in the repository.</p><p> </p><p>This work is funded by Digital Futures in the project EO-AI4GlobalChange. The computations were enabled by resources provided by the National Academic Infrastructure for Supercomputing in Sweden (NAISS) at C3SE partially funded by the Swedish Research Council through grant agreement no. 2022-06725.</p>
Figure 6. Performance plot for NASDAQ index (RNN)-Comparative study of Financial Time Series Prediction by Artificial Neural Network with Gradient Descent Learning
<p>Figure 5 we can see that the mse curve reaches the In performance goal but it does not<br> decrease in that good manner,but in Figure 6 the mse is reduces widely. By analyzing all these<br> results one can say that RNN is better choice than Feedforward MLP in prediction purpose.</p>
Figure 5. Performance plot for NASDAQ index (MLP)-Comparative study of Financial Time Series Prediction by Artificial Neural Network with Gradient Descent Learning
<p>Figure 5 we can see that the mse curve reaches the In performance goal but it does not<br> decrease in that good manner,but in Figure 6 the mse is reduces widely. By analyzing all these<br> results one can say that RNN is better choice than Feedforward MLP in prediction purpose.</p>
Figure 3. Regression plot for NASDAQ index (MLP)-Comparative study of Financial Time Series Prediction by Artificial Neural Network with Gradient Descent Learning
<p>Figure 3 depicts the regression plot for the feedforward MLP network, analyzing it we can<br> say that Y=T regression is not so good.</p>
Figure 4. Regression plot for NASDAQ index (RNN)-Comparative study of Financial Time Series Prediction by Artificial Neural Network with Gradient Descent Learning
<p>Figure 4,depicts the regression plot for the Timedelay RNN network, analyzing it we can<br> say that Y=T regression is totally fit.<br> This paper also comprises of comparative study of performance(mse) plot of both network.</p>
Figure 2.Flow Chart for Data preprocessing & Training-Comparative study of Financial Time Series Prediction by Artificial Neural Network with Gradient Descent Learning
<p>Methodology<br> This paper develops an ANN based comparative predictive model for NASDAQ stock<br> prediction. The first ANN model is developed with Multi-Layer Feed forward Network<br> Architecture & the second model is developed with Recurrent Neural Network Architecture. In this<br> paper gradient descent based back propagation learning algorithm is used for the supervised<br> learning of the predictive network.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 18. Drawing of the time series for males and females of primary stage students and its prediction)
<p>Note that the Tabulated value equals 3.841 while the Q value is less than Tabulated value, so it takes the Null Hypothesis which manifests that the emptiness of the evaluated model out of the contrast in accordance trouble. It's possible to notice that the two parameters functions (Autocorrelation and Partial correlation Functions) of the residues for male females primary stage, in which the residues value is located within confidence interval limits, which means the residues series is random and the evaluated model is good and convenient as it is presented.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 17. Drawing of the time series for females of primary stage students and its prediction
<p>It's possible to notice that the two parameters functions (Autocorrelation and Partial correlation Functions) of the residues for male females primary stage, in which the residues value is located within confidence interval limits, which means the residues series is random and the evaluated model is good and convenient as it is presented.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 16. Drawing of the time series for males of primary stage students and its prediction
<p>It's possible to notice that the two parameters functions (Autocorrelation and Partial correlation Functions) of the residues for male females primary stage, in which the residues value is located within confidence interval limits, which means the residues series is random and the evaluated model is good and convenient as it is presented.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 11. Drawing the time series for males and females primary stage after the First difference
<p>We use the Unit Radix Dickey-Fuller Test to ensure the series’ stability. The results are: Dickey-Fuller Test Estimated Value = 0.369693, Statistic Test =1.01829, P-Value=0.9194 We notice from the values above P-Value = 0.9194 on the abstract level of 0.05 which leads to accepting the Null Hypothesis and refusing the Alternative Hypothesis (Existence of a Radix Unit) implies that the time series is instable. By taking the first difference, we notice that the stability of the time series has been achieved. See Figure 11.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 9. Drawing the time series for males and females primary stage students
<p>The Unit Radix Dickey-Fuller Test is used to ensure the series’ stability. The results are: Dickey-Fuller Test Estimated Value = 0.736458 , Statistic Test = 0.380545 , P-Value = 0.794 We notice from the values above P-Value = 0.794on the abstract level of 0.05 which leads to refusing the Null Hypothesis and accepting the Alternative Hypothesis ( The Nonexistence of a Radix Unit) implies that the time series is stable. Figure (9) represents the time series of females and males in the primary stage students.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 5. Drawing the time series for primary stage female students
<p>We use the Unit Radix Dickey-Fuller Test to assure the series’ stability. The results are: Dickey-Fuller Test Estimated Value = 0.233403, Statistic Test = 0.125769, P-Value = 0.6405 The values above P-Value = 0.6405 is noted on the abstract level of 0.05 which leads to refusing the Null Hypothesis and accepting the Alternative Hypothesis (The Nonexistence of a Radix Unit) implies that the time series is stable. Figure 5 represents the Time series of Female Primary Stage Students.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 3. Drawing the time series for Males primary stage students after getting the first difference
<p>The Unit Radix Dickey-Fuller Test is used to assure the series’ stability. The results are: Dickey-Fuller Test Estimated Value = 0.276151, Statistic Test = 0.87796, P-Value = 0.8984 We get to notice from the values above P-Value = 0.8984 on the abstract level of 0.05 which leads to accepting the Null Hypothesis and refusing the Alternative Hypothesis (Existence of a Radix Unit) implies that the time series is instable. By taking the first difference, it is observed that the stability of the Time Series has been accomplished . See figure 3.</p>
The Prediction of the Rate of the Dropout of the Primary Schools Students by Using the Genetic Algorithm-Figure 1. Drawing the time series for males primary stage
<p>After collecting all the students’ dropout proportion for both males and females in the<br> primary stage, the first step of the Box-Jenkins is to draw the time chain data to understand the<br> chain's attitude.</p>
Soil organic carbon models need independent time-series validation for reliable prediction
<p>Supplementary Data 1 to the paper: Soil organic carbon models need independent time-series validation for reliable prediction</p> <p>By: Le Noë, J., Manzoni, S., Abramoff, R.Z., Bölscher, T., Bruni, E., Cardinael, R., Ciais, P., Chenu, C., Clivot, H., Derrien, D., Ferchaud, F., Garnier, P., Goll, D., Lashermes, G., Martin, M.P., Rasse, D., Rees, F., Sainte-Marie, J., Salmon, E., Schiedung, M., Schimel, J., Wieder, W.R., Abiven, S., Barré, P., Cécillon, L., Guenet, B.</p>
Establishing diversity in synthetic time series for prediction performance evaluation
<p>This dataset enables practitioners to evaluate their time series prediction algorithms on various types of time series</p>
ScienceDex guides
Understand access before you commit
These curated guides explain access requirements, typical timelines, costs, and reuse considerations for widely used research datasets.
Allen Brain Atlas
Allen Brain Atlas is an Allen Institute collection of brain map atlases, datasets, APIs, and analysis tools covering mouse, human, and non-human primate brain resources.
Annotated Behaviour and Observability Dataset (ABODe)
ABODe is a University of Edinburgh DataShare dataset for behavior classification in group-housed mice using home-cage video, identities, bounding boxes, ground-plate positions, and annotator labels.
DANDI Archive for NWB datasets
DANDI is a BRAIN Initiative archive for publishing and sharing neurophysiology data, including electrophysiology, optophysiology, and behavioral data packaged as NWB and related standards.
International Brain Laboratory public data
The International Brain Laboratory public data releases expose standardized mouse decision-making experiments, including Neuropixels recordings, widefield calcium imaging, behavior, and session metadata accessed through the ONE API.
OpenNeuro
OpenNeuro is a free, open platform for sharing neuroimaging datasets, with public search, dataset pages, and download paths for web, S3, DataLad, and the OpenNeuro CLI.