Find research datasets worth reusing

Data for Streamflow Prediction: Comparison of SWAT vs. Random Forest Models in Diverse Catchments

<p>This study introduces a time-lag-informed Random Forest (RF) framework for streamflow time series prediction across diverse catchments, and compares its results against SWAT predictions. We found strong evidence of RF's better performance by adding historical flows and time-lags for meteorological values over using only actual meteorological values. On a daily scale, RF demonstrated robust performance (Nash–Sutcliffe efficiency [<em>NSE</em>] > 0.5), whereas SWAT generally yielded unsatisfactory results (<em>NSE</em> < 0.5) and tended to overestimate daily streamflow by up to 27% (<em>PBIAS</em>). However, SWAT provided better monthly predictions, particularly in catchments with irregular flow patterns. Although both models faced challenges in predicting peak flows in snow-influenced catchments, RF outperformed SWAT in an arid catchment. RF also exhibited a notable advantage over SWAT in terms of computational efficiency. Overall, RF is a good choice for daily predictions with limited data, whereas SWAT is preferable for monthly predictions and understanding hydrological processes in depth.</p> <p>This repository contains the input data used for building the RF and SWAT models and the files describing the modeling results.</p> <p>The corresponding Zenodo code repository is available at <a href="../doi/10.5281/zenodo.11064973" target="_blank" rel="noopener">https://zenodo.org/doi/10.5281/zenodo.11064973</a>.</p>

The Kling-Gupta efficiency (KGE) and latent variable data to analyze the dynamic changes and driving mechanisms of predictability

Open the record for dataset details and reuse information.

Universal Rapid Weather Prediction Model (Sonagi Model) Korea Peninsula 5 Days Forecast Result (Pressure Isobar Map)

<p>Universal Rapid Weather Prediction Model (Sonagi Model) Korea Peninsula 5 Days Forecast Result (Pressure Isobar Map)</p> <p>Each file has its altitude in front of the name of the file, and by each isobar in map directs the air current heading higher or lower altitude. And rest of the name follows the target ed UTC time. Generally, iso-temperature lines are used to indicate where air flows, but it was simulatable in Sonagi model to where air is heading by pressure, so that pressure isobar is used to indicate where the air flows.</p> <p>Input data for the prediction in Universal Rapid Weather Prediction Model (Sonagi Model) is originated from GK2A satelite of Korea Meteorological Administration. (https://apihub.kma.go.kr/) For sharing the original prediction data, contact me at somehowme@gmail.com or flyingtext@nate.com (Prediction netCDF4 files are almost 16GB in sum total.)</p>

Universal Rapid Weather Prediction Model (Sonagi Model) Pacific Ocean 5 Days Forecast Result (Pressure Isobar Map)

<p>Universal Rapid Weather Prediction Model (Sonagi Model) Pacific Ocean 5 Days Forecast Result (Pressure Isobar Map)</p> <p>Each file has its altitude in front of the name of the file, and by each isobar in map directs the air current heading higher or lower altitude. And rest of the name follows the target ed UTC time. Generally, iso-temperature lines are used to indicate where air flows, but it was simulatable in Sonagi model to where air is heading by pressure, so that pressure isobar is used to indicate where the air flows.</p> <p>Input data for the prediction in Universal Rapid Weather Prediction Model (Sonagi Model) is originated from GK2A satelite of Korea Meteorological Administration. (https://apihub.kma.go.kr/) For sharing the original prediction data, contact me at somehowme@gmail.com or flyingtext@nate.com (Prediction netCDF4 files are almost 16GB in sum total.)</p>

Code Generation for Predicting the Radium Activity Equivalent

Open the record for dataset details and reuse information.

Input data for predicted sedimentary environments on the Norwegian continental margin

<p>Input and output data relating to R workflow for predicting sedimentary environments on the Norwegian continental margin (https://github.com/diesing-ngu/SedEnv). The following files are included:</p><p><strong>SedEnv_4km_MaxCombArea_point_20230622.shp</strong> - Point shapefile of the response data (substrate type). Note that these data points were derived from mapped products and are not sample points as such. </p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif</strong> - Georeferenced TIFF-file of predicted substrate classes. Used to update the area of interest (exclude areas mapped as Rock and boulders).</p><p><strong>mud_2023-06-30.tif </strong>- Georeferenced TIFF-file of predicted mud content. Used as an additional predctor.</p>

Predicted mud content on the Norwegian continental margin

<p>Output data relating to R workflow for predicting mud content on the Norwegian continental margin (https://github.com/diesing-ngu/GSMgrids). The following files are included:</p><p><strong>alrM_2023-10-31.tif </strong>- Georeferenced Tiff-file of the predicted additive log-ratio model</p><p><strong>alrM_aoa2023-10-31.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p><p><strong>mud_2023-06-30.tif </strong>- Georeferenced Tiff-file of the mud content</p>

Predicted organic carbon content on the Norwegian continental margin

<p>Output data relating to R workflow for predicting organic carbon content on the Norwegian continental margin (https://github.com/diesing-ngu/TOC). The following files are included:</p> <p><strong>OC0-10cm_median_2024-02-16.tif</strong> - Georeferenced Tiff-file of the predicted organic carbon content in the upper 10 cm of the sediment column</p> <p><strong>OC0-10cm_PI90_2024-02-16.tif</strong> - Georeferenced Tiff-file of the 90% prediction interval. Can be used as a measure of uncertainty.</p> <p><strong>OC0-10cm_AOA_2024-02-16.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p> <p><strong>OC0-10cm_AOA_2024-02-16.shp</strong> - Same as above but as polygon shapefile</p>

Input data for predicted sediment accumulation rates on the Norwegian continental margin

<p>Input data relating to R workflow for predicting sediment accumulation rates on the Norwegian continental margin (https://github.com/diesing-ngu/SedRates). The following files are included:</p><p><strong>norway_sar_2023-08-28.csv</strong> - Data on sediment accumulation rates from the <a href="https://doi.org/10.5194/essd-15-4105-2023">MOSAIC v2.0</a> database</p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif</strong> - Georeferenced TIFF-file of predicted substrate classes. Used to update the area of interest (exclude areas mapped as Rock and boulders).</p><p><strong>mud_2023-06-30.tif </strong>- Georeferenced TIFF-file of predicted mud content. Used as an additional predctor.</p><p><strong>SedEnv3_probabilities_2023-07-01.tif </strong>- Georeferenced Tiff-file of the prediction probabilities of the predicted sedimentary environments. Used as additional predctors.</p><p><strong>SedEnv3_max_probabilities_2023-07-01.tif </strong>- Georeferenced Tiff-file of the maximum probabilities, i.e., the probability of the class that was mapped.</p><p><strong>SedEnv3_AOA_2023-07-01.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p>

Input data for predicted dry bulk densities on the Norwegian continental margin

<p>Input data relating to R workflow for predicting dry bulk densities on the Norwegian continental margin (https://github.com/diesing-ngu/DBD). The following files are included:</p><p><strong>DBD_2023-07-21.csv </strong>- Data on dry bulk densities in surface sediments</p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif</strong> - Georeferenced TIFF-file of predicted substrate classes. Used to update the area of interest (exclude areas mapped as Rock and boulders).</p><p><strong>GrainSizeReg_folk8_probabilities_2023-06-28.tif</strong> - Georeferenced TIFF-file of the prediction probabilities of the predicted substrate classes. Used as additional predctors.</p>

Input data for predicted substrate types on the Norwegian continental margin

<p>Input data relating to R workflow for predicting substrate types on the Norwegian continental margin (https://github.com/diesing-ngu/GrainSizeReg). The following files are included:</p><p><strong>AoI_Harris_mod </strong>- Polygon shapefile delimiting the area of interest</p><p><strong>GrainSize_4km_MaxCombArea_folk8_point_20230628</strong> - Point shapefile of the response data (substrate type). Note that these data points were derived from mapped products and are not sample points as such. </p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p>

Predicted sediment accumulation rates on the Norwegian continental margin

<p>Output data relating to R workflow for predicting sediment accumulation rates on the Norwegian continental margin (https://github.com/diesing-ngu/SedRates). The following files are included:</p> <p><strong>SAR_median_2024-02-15.tif</strong> - Georeferenced Tiff-file of the oredicted sediment accumulation rates</p> <p><strong>SAR_PI90_2024-02-15.tif</strong> - Georeferenced Tiff-file of the 90% prediction interval. Can be used as a measure of uncertainty.</p> <p><strong>SAR_AOA_2024-02-15.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p> <p><strong>SAR_AOA_2024-02-15.shp</strong> - Same as above but as polygon shapefile</p>

Predicted sedimentary environments on the Norwegian continental margin

<p>Output data relating to R workflow for predicting substrate types on the Norwegian continental margin (https://github.com/diesing-ngu/SedEnv). The following files are included:</p><p><strong>SedEnv3_classes_2023-07-01.tif </strong>- Georeferenced Tiff-file of the predicted sedimentary environments</p><p><strong>SedEnv3_probabilities_2023-07-01.tif </strong>- Georeferenced Tiff-file of the prediction probabilities of the predicted sedimentary environments</p><p><strong>SedEnv3_max_probabilities_2023-07-01.tif </strong>- Georeferenced Tiff-file of the maximum probabilities, i.e., the probability of the class that was mapped. Can be used as an indicator of map confidence.</p><p><strong>SedEnv3_AOA_2023-07-01.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p><p><strong>SedEnv3_AOA_2023-07-01.shp</strong> - Same as above but as polygon shapefile</p>

Predicted dry bulk densities on the Norwegian continental margin

<p>Output data relating to R workflow for predicting dry bulk densities on the Norwegian continental margin (https://github.com/diesing-ngu/DBD). The following files are included:</p> <p><strong>DBD0-10cm_median_2024-02-16.tif</strong> - Georeferenced Tiff-file of the Predicted dry bulk densities in the upper 10 cm of the sediment column</p> <p><strong>DBD0-10cm_PI90_2024-02-16.tif</strong> - Georeferenced Tiff-file of the 90% prediction interval. Can be used as a measure of uncertainty.</p> <p><strong>DBD0-10cm_AOA_2024-02-16.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p> <p><strong>DBD0-10cm_AOA_2024-02-16.shp</strong> - Same as above but as polygon shapefile</p>

Input data for predicted organic carbon content on the Norwegian continental margin

<p>Input data relating to R workflow for predicting organic carbon content on the Norwegian continental margin (https://github.com/diesing-ngu/TOC). The following files are included:</p><p><strong>mosaic_2023-04-21.csv</strong> - Data on organic carbon content in surface sediments from the <a href="https://doi.org/10.5194/essd-15-4105-2023">MOSAIC v2.0</a> database</p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif</strong> - Georeferenced TIFF-file of predicted substrate classes. Used to update the area of interest (exclude areas mapped as Rock and boulders).</p><p><strong>mud_2023-06-30.tif </strong>- Georeferenced TIFF-file of predicted mud content. Used as an additional predctor.</p>

Input data for predicted mud content on the Norwegian continental margin

<p>Input data relating to R workflow for predicting mud content on the Norwegian continental margin (https://github.com/diesing-ngu/GSMgrids). The following files are included:</p><p><strong>gsm_data.csv</strong> - Data on gravel, sand and mud of surface sediments</p><p><strong>MGObsPkt_150_160_170.shp </strong>- Point shapefile of categorical samples</p><p><strong>predictors_ngb.tif </strong>- Multi-band georeferenced TIFF-file of predictor variables</p><p><strong>predictors_description</strong>.txt - Information on variables stored in predictor_ngb.tif including units, statistics, time period and sources.</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif</strong> - Georeferenced TIFF-file of predicted substrate classes. Used to update the area of interest (exclude areas mapped as Rock and boulders).</p><p><strong>GrainSizeReg_folk8_probabilities_2023-06-28.tif</strong> - Georeferenced TIFF-file of the prediction probabilities of the predicted substrate classes. Used as additional predctors.</p>

Predicted substrate types on the Norwegian continental margin

<p>Output data relating to R workflow for predicting substrate types on the Norwegian continental margin (https://github.com/diesing-ngu/GrainSizeReg). The following files are included:</p><p><strong>GrainSizeReg_folk8_classes_2023-06-28.tif </strong>- Georeferenced Tiff-file of the predicted substrate classes</p><p><strong>GrainSizeReg_folk8_probabilities_2023-06-28.tif </strong>- Georeferenced Tiff-file of the prediction probabilities of the predicted substrate classes</p><p><strong>GrainSizeReg_folk8_max_probabilities_2023-06-28.tif </strong>- Georeferenced Tiff-file of the maximum probabilities, i.e., the probability of the class that was mapped. Can be used as an indicator of map confidence.</p><p><strong>GrainSizeReg_folk8_AOA_2023-06-28.tif </strong>- Georeferenced Tiff-file of the area of applicability of the model <a href="https://doi.org/10.1111/2041-210X.13650">(Meyer & Pebesma, 2021)</a></p><p><strong>GrainSizeReg_folk8_AOA_2023-06-28.shp</strong> - Same as above but as polygon shapefile</p>

opencc-by-4.0May 2024View details →

Predictive Modeling of Bearing Degradation: LSTM Neural Networks for Uncertainty Quantification

<p>These MATLAB codes are part of a research project focused on predicting bearing degradation through vibration measurements. The codes implement LSTM (Long Short-Term Memory) neural network models trained under different objectives, including uncertainty quantification and RMSE (Root Mean Square Error) minimization. The objective of the research is to compare the performance of these models in predicting bearing health and assessing the associated uncertainty.</p> <p><strong>Note:</strong> The current codes are under embargo access as the corresponding paper has been submitted to the ESCA 11 conference. The codes will be made openly accessible upon acceptance of the paper and during the presentation dates. Please cite our paper when using these codes.</p>

opencc-by-4.0May 2024View details →

Enhancing Change Impact Prediction by Integrating Evolutionary Coupling with Software Change Relationships

<div> <div><strong>Replication Package For Paper: Enhancing Change Impact Prediction by Integrating Evolutionary Coupling with Software Change Relationships</strong></div> <br> <div>IMPER is a change <strong>IM</strong>pact <strong>P</strong>rediction method that integrates <strong>E</strong>volutionary coupling with software change <strong>R</strong>elationships. The approach consists of following four main steps:</div> <br> <div><em>Step 1</em>: Relevant Transaction Filtering. Filtering the relevant transaction set based on the given query $<em>Q</em>$.</div> <div><em>Step 2</em>: Association Rule Generation. Generating association rules and obtaining the impact candidate set.</div> <div><em>Step 3</em>: Spatial-Temporal Window Processing. Considering the temporal and spatial change relationships to obtain the impact candidate set.</div> <div><em>Step 4</em>: Candidates Integrating and Ranking. Making a union of these two sets to find the final impact set.</div> <br> <div>The repo provides files for the paper replication. It includes::</div> </div> <div> </div> <div><strong>git_repo.zi</strong>p: Git repository used in the study.</div> <div><em><strong>***-rec-database-1107.tar.gz</strong>:</em> Neo4j dataset containing entity-level software relationship graphs and change information (changed classes, member member methods, member member variables, and files) from over 19,003 versions of six well-maintained projects.</div> <div><strong><em>code_entiy_genealogy</em></strong>: Data for code entity genealogy.<br> <div><em><strong>imper_jar.zip</strong>:</em>  Tool for the study, including: 1) Generating qualified queries; 2) Change prediction using four approaches (ROSE, TARMAQ, IMPER, IMPER-ST); 3) Evaluation. The tool outputs evaluation results for each query and average metrics (MAP, Precision, Recall, and F1-score).</div> <div><strong><em>evaluation_metrics_details.zip</em></strong>: Detailed evaluation metric results for different projects.</div> <div><strong><em>Imper_evaluation_results.xlsx</em></strong>: Summary of evaluation metric results for all projects.</div> <div> </div> <br> <div>Users are suggested to refer to the research paper for more detials.</div> <br> <div>This project is still in an early stage of development, and we are improving the implementation and documentation. Please feel free to contact ******(author) for the replication package.</div> <br><strong>Environment</strong><br> <div>- java 11+</div> <div>- <a href="https://neo4j.com/download-center/#community">neo4j 4.2</a> </div> <div>- memory 64G+</div> <br><strong>Quick Start</strong>:<br>To begin our study, follow these steps. Note that only one project can be analyzed at a time.</div> <div><br><strong><em>Step 1: Database Setup</em></strong></div> <div> </div> <div>Copy the relevant database to the folder `data` (in the root path of the Neo4j database), copy the plugins to the plugin folder, and open the database using Neo4j 4.2 (Recommended).</div> <div><br><strong><em>Step 2: Configure application-dev.yml</em></strong></div> <div> </div> <div>Adjust the application-dev.yml file in the folder `imper_jar` to match your configuration. If you're using an existing database, simply configure the database address, username, and password. <div> <p> </p> </div> </div> <div><strong><em>Step 3:  Run imper-study.jar</em></strong></div> <div><br>Command: `<em>java -jar imper-study.jar -imp -config ./application-dev.yml -rd /home/user/code_rec -pj commons-io -fi commons-io_file(Map)_1031.txt -mi commons-io_functionInfo(Map).txt -tw 35 -mc 20 -od /home/user/code_rec</em>`.</div> <div><br> <div>-imp:  Used to run IMPER</div> <div>-config:  Path of `<em>application-dev.yml</em>`.</div> <div>-rd:  Root path of code entiy genealogy, e.g., `<em>/home/user/code_rec</em>`.</div> <div>-pj:  Project name, e.g., `<em>commons-io</em>`.</div> <div>-fi:  the code entiy genealogy (file), e.g., `<em>commons-io_file(Map)_1031.txt</em>` (the absolute path is `<em>/home/user/code_rec/commons-io/commons-io_file(Map)_1031.txt</em>`).</div> <div>-mi:  the code entiy genealogy (method), e.g., `<em>commons-io_functionInfo(Map).txt</em>` (the absolute path is `<em>/home/user/code_rec/commons-io/commons-io_functionInfo(Map).txt</em>`).</div> <div>-tw:  Time window (half), e.g., <em>-tw 35 (the time window size is 70).</em></div> <div>-mc:  Maxium commit size in the time window.</div> <div>-od:  Output root path for evaluation results (MAP, Precison, Recall and F1-score), e.g., `<em>/home/user/code_rec</em>`.</div> </div>