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.
1,549
datasets available to search
ShareScore release 0.9.0
Dataset results
1,549 results for “benchmarks”
Benchmark for the evaluation of named entity recognition over ancient documents
<p>The dataset consists of a multilingual noisy corpora for named entity recognition (NER).<br> The noisy versions are simulated from the CoNLL-02 (Spanish and Dutch) and CoNLL-03 (English) NER corpora.<br> The original collections are re-OCRed and four types of noises at two different levels are added in order to simulate various OCR output.</p> <p>More precisely, we first extracted raw texts and converted them into images. These images have been contaminated by adding some common noises when using a scanner. We further extract OCRed data using tesseract open source<br> OCR engine v-3.04.01. Consequently to the image noise insertions, OCRed data contains degradations. Original and noisy texts are finally aligned.</p> <p>This archive contains three folders (one per language). The folders contain the degraded images, the noisy texts extracted by the OCR and their aligned version with clean data.</p> <p>These are the supplementary materials for the TPDL 2020 paper <a href="https://zenodo.org/record/4734376#.YJKAcKE6-Uk">Assessing and minimizing the impact of OCR quality on named entity recognition</a>. If you end up using whole or parts of this resource,<br> please cite this paper:</p> <pre><code>@InProceedings{10.1007/978-3-030-54956-5_7, author="Hamdi, Ahmed and Jean-Caurant, Axel and Sid{\`e}re, Nicolas and Coustaty, Micka{\"e}l and Doucet, Antoine", editor="Hall, Mark and Mer{\v{c}}un, Tanja and Risse, Thomas and Duchateau, Fabien", title="Assessing and Minimizing the Impact of OCR Quality on Named Entity Recognition", booktitle="Digital Libraries for Open Knowledge", year="2020", publisher="Springer International Publishing", address="Cham", pages="87--101", isbn="978-3-030-54956-5" }</code></pre> <p><strong>Acknowledgments</strong><br> This work has been supported by the European Union's Horizon 2020 research and innovation programme under grant 770299 [NewsEye](https://www.newseye.eu/).</p>
Heteroplasmy Benchmark Dataset - mitochondrial DNA mixture model - MiSeq - U5-H1-M1-M2-M3-M4-M5 - FASTQ
<p>mtDNA mixture model of 2 mtDNA sequences belonging to haplogroups U5 and H1. Run on Illumina MiSeq with 3 different polymerases (Clontech, Herculase, NEB Taq), and different DNA extraction protocols - Paired-end Fastq files</p> <p>M1 = Mixture 1:2 i.e. 50%</p> <p>M2 = Mixture 1:10 i.e. 10%</p> <p>M3 = Mixture 1:50 i.e. 2%</p> <p>M4 = Mixture 1:100 i.e. 1%</p> <p>M5 = Mixture 1:200 i.e. 0.5%</p>
CoDEx: A Comprehensive Knowledge Graph Completion Benchmark
<p>This repository hosts the <strong>relational-only part</strong> of the CoDEx benchmark, which was presented at the EMNLP 2020 conference. You can access the paper <a href="https://www.aclweb.org/anthology/2020.emnlp-main.669.pdf">here</a> and the full dataset, including text and pretrained models, <a href="https://bit.ly/2EPbrJs">on GitHub</a>.</p> <p>Abstract:</p> <p><em>We present CoDEx, a set of knowledge graph completion datasets extracted from Wikidata and Wikipedia that improve upon existing knowledge graph completion benchmarks in scope and level of difficulty. In terms of scope, CoDEx comprises three knowledge graphs varying in size and structure, multilingual descriptions of entities and relations, and tens of thousands of hard negative triples that are plausible but verified to be false. To characterize CoDEx, we contribute thorough empirical analyses and benchmarking experiments. First, we analyze each CoDEx dataset in terms of logical relation patterns. Next, we report baseline link prediction and triple classification results on CoDEx for five extensively tuned embedding models. Finally, we differentiate CoDEx from the popular FB15K-237 knowledge graph completion dataset by showing that CoDEx covers more diverse and interpretable content, and is a more difficult link prediction benchmark. Data, code, and pretrained models are available <a href="https://bit.ly/2EPbrJs">here</a>.</em></p>
Supplementary data to *Benchmarking of numerical integration methods for ODE models of biological systems*
<p>This archive contains supplementary data and code for the manuscript <strong>Benchmarking of numerical integration methods for ODE models of biological systems </strong>by<strong> Städter et al. 2020</strong>. It contains</p> <ul> <li>scripts to automatically download and install all required packages and models,</li> <li>scripts to compile the models and to perform the study,</li> <li>value files containing all data underlying the analyses in the manuscript,</li> <li>scripts to generate the manuscript figures.</li> </ul> <p>There is a <strong>README.md </strong>file with further information, in particular on what scripts to execute to reproduce the study.</p>
FIA Benchmark Biomass Estimates for sub-regions of the United States
<p>This attached GIS (available in both geodatabase and kml formats) contains all of the hexagon-level estimates described by Menlove and Healey's Technical Note: "A comprehensive forest biomass dataset for the US allows customized validation of remotely sensed biomass estimates." Version 1.2: 28 November, 2020</p>
Acceleration response of a benchmark 3 Story Structure (SAC) with MR damper under earthquake excitation
<p>This file contains acceleration response of benchmark 3 story structure equipped with MR damper under the Chuetsu-Oki earthquake in Niigata prefecture, Japan, which occurred on July 16th, 2007. Acceleration response of each story is separated and saved in separate text files. Text file contains four columns that the first one has been devoted to time series, while the remained (second, third, fourth) columns present acceleration responses in three perpendicular axes, X, Y, Z, respectively.</p>
MCMC simulation data for AutoGibbs.jl benchmarks
<p>Chains and diagnostic data for the test models of <a href="https://github.com/phipsgabler/AutoGibbs.jl">AutoGibbs.jl.</a> Includes four CSV files for each of the models: GMM, HMM, and IMM.</p> <p> </p> <p>Files are all in long format. The files for chains, diagnostics, and sampling times all contain "model", "discrete_algorithm" (AG or PG), "continuous_algorithm" (always HMC), "particles" (always 100), "data_size" (number of observations: 10, 25, or 50), and "repetition" (index of chain; there are 10 chains per parameter combination, except for HMC, there the benchmark has been killed somewhere after the eight chain). In addition to that, the sampling time files contain a column "samping_time" (in seconds), and the diagnostic files have columns "parameter" (name of the random variable), "diagnostic" (ESS or R_hat), and "value" with chain diagnostic values. Chain files have "parameter", "step", and "value" for the sampled values at each step of the chain.</p> <p> </p> <p>The compile time measurements consist just of columns "model", "data_size", "repetition", "compilation_time" (in seconds).</p> <p> </p> <p>Note that the "parameter" column in the HMM data has been manually adjusted; see the file plotting.jl in the AutoGibbs.jl repository.</p>
Container Registry Benchmark experiments measurements and trace workload samples
<p>Measurements for experiments using Container Registry Benchmark, CReB. 4 experiments: Long running, small experiment stress mode, small experiment delay mode, and large workload experiment.</p> <p> </p> <p>Structure:</p> <ol> <li><strong>full-measurements-long-running-pull.csv : </strong>measurements for long running pull experiment</li> <li><strong>full-measurements-long-running-push.csv: </strong>measurements for long running push experiment</li> <li><strong>result-bug-analysis.zip: </strong>results from bug analysis of trace replayer</li> <li><strong>results-1hr-experiment.zip: </strong>measurements for the large experiment (4 registries)</li> <li><strong>results-small-delay.zip: </strong>measurements for the delay mode, small experiment with real workload</li> <li><strong>results-small-stress.zip: </strong>measurements for the stress mode, small experiment with real workload</li> <li><strong>traces.zip: </strong>traces used for pen-and-paper experiment, 1 hour sample, and the trace used for small experiment (selected are first 405 requests)</li> </ol>
Industrial Benchmark Dataset for Customer Escalation Prediction
<p>This is a real-world industrial benchmark dataset from a major medical device manufacturer for the prediction of customer escalations. The dataset contains features derived from IoT (machine log) and enterprise data including labels for escalation from a fleet of thousands of customers of high-end medical devices. </p> <p>The dataset accompanies the publication "System Design for a Data-driven and Explainable Customer Sentiment Monitor" (submitted). We provide an anonymized version of data collected over a period of two years.</p> <p>The dataset should fuel the research and development of new machine learning algorithms to better cope with real-world data challenges including sparse and noisy labels, and concept drifts. Additional challenges is the optimal fusion of enterprise and log based features for the prediction task. Thereby, interpretability of designed prediction models should be ensured in order to have practical relevancy. </p> <p><strong>Supporting software</strong></p> <p>Kindly use the corresponding <a href="https://github.com/annguy/customer-sentiment-monitor">GitHub repository</a> (https://github.com/annguy/customer-sentiment-monitor) to design and benchmark your algorithms. </p> <p> </p> <p><strong>Citation and Contact</strong><br> </p> <p>If you use this dataset please cite the following publication:</p> <p><br> </p> <pre><code>@ARTICLE{9520354, author={Nguyen, An and Foerstel, Stefan and Kittler, Thomas and Kurzyukov, Andrey and Schwinn, Leo and Zanca, Dario and Hipp, Tobias and Jun, Sun Da and Schrapp, Michael and Rothgang, Eva and Eskofier, Bjoern}, journal={IEEE Access}, title={System Design for a Data-Driven and Explainable Customer Sentiment Monitor Using IoT and Enterprise Data}, year={2021}, volume={9}, number={}, pages={117140-117152}, doi={10.1109/ACCESS.2021.3106791}}</code></pre> <p> </p> <p>If you would like to get in touch, please contact an.nguyen@fau.de.<br> </p>
Time Series used in the Forecasting Benchmark
<p>This data set contains the time series used in Libra (GitHub: <a href="https://github.com/DescartesResearch/ForecastBenchmark">https://github.com/DescartesResearch/ForecastBenchmark</a> ; CodeOcean: <a href="https://doi.org/10.24433/CO.3240518.v1">https://doi.org/10.24433/CO.3240518.v1</a>). Libra is a forecasting benchmark that automatically evaluates and ranks forecasting methods based on their performance in a diverse set of evaluation scenarios. The benchmark comprises four different use cases, each covering 100 heterogeneous time series taken from different domains.</p>
Benchmark maps and assignments for multi-agent path finding
<p><strong>The dataset</strong></p> <p>The dataset is produced within the SafeLog project and it is used for benchmarking of multi-agent path planning algorithms. Specifically, the dataset consists of a set of 21 maps with increasing density and a set of 500 random assignments, each for a group of 100 agents for planning on each of the maps.</p> <p>All of the maps, in the form of a graph G = {V, E}, are built on the same set of 400 vertices V. The sets of edges Ej, where j ∈ (0; 20), in the maps then form a set ranging from a spanning tree to a mostly 4-connected graph. These maps were created by generating a complete square graph with the size of 20*20 vertices. The graph was then simplified to a spanning tree, and, finally, approximately 50 random edges from the complete graph were added 20 times, to create the set of 21 maps of density ranging from 800 to 1500 edges in the graph. <br> </p> <p><strong>Content and format</strong></p> <p>The following files are included in the dataset</p> <p><em>test_nodes.txt </em>- 400 nodes of a 20*20 square map in the form "id x y"<br> <em>testAssignment.txt</em> - 50499 random pairs of nodes ids from test_nodes.txt<br> <em>test_edgesX.txt</em> - pairs of adjacent nodes ids from test_nodes.txt forming edges<br> - X = 0 - tree<br> - X = 20 - full graph<br> - created starting at a full graph and repeatedly erasing edges until a tree remains</p> <p>To illustrate the maps in the dataset, we provide three images (1008.png, 1190.png, and 1350.png) showing maps with 1008 (1190, 1350) edges. </p> <p><strong>Citation</strong></p> <p>If you use the dataset, please cite:</p> <p>[1] Hvězda, J., Rybecký, T., Kulich, M., and Přeučil, L. (2018). <a href="https://doi.org/10.1109/ITSC.2018.8569712"><em>Context-Aware Route Planning for Automated Warehouses</em></a>. Proceedings of 2018 21st International Conference on Intelligent Transportation Systems (ITSC).</p> <p><em>@inproceedings{Hvezda18itsc,<br> author = {Hvězda, Jakub and Rybecký, Tomáš and Kulich, Miroslav and Přeučil, Libor},<br> title = {Context-Aware Route Planning for Automated Warehouses},<br> booktitle = {Proceedings of 2018 21st International Conference on Intelligent Transportation Systems (ITSC)},<br> publisher = {IEEE Intelligent Transportation Systems Society},<br> address = {Maui},<br> year = {2018},<br> doi = {10.1109/ITSC.2018.8569712},<br> }</em></p> <p> </p> <p>[2] Hvězda, J., Kulich, M., and Přeučil, L. (2019). <a href="https://link.springer.com/chapter/10.1007/978-3-030-31993-9_18">On Randomized Searching for Multi-robot Coordination</a>. In: Gusikhin O., Madani K. (eds) Informatics in Control, Automation and Robotics. ICINCO 2018. Lecture Notes in Electrical Engineering, vol 613. Springer, Cham.</p> <p><em>@inbook{Hvezda19springer,<br> author = {Hvězda, Jakub and Kulich, Miroslav and Přeučil, Libor},<br> title = {On Randomized Searching for Multi-robot Coordination},<br> booktitle = {Informatics in Control, Automation and Robotics},<br> publisher = {Springer},<br> address = {Cham, CH},<br> year = {2019},<br> series = {Lecture Notes in Electrical Engineering},<br> language = {English},<br> url = {https://link.springer.com/chapter/10.1007/978-3-030-31993-9_18},<br> doi = {10.1007/978-3-030-31993-9},<br> }</em></p> <p><br> [3] Hvězda, J., Kulich, M., and Přeučil, L. (2018). <a href="http://www.scitepress.org/PublicationsDetail.aspx?ID=ppwUqsGaX18=\&t=1">Improved Discrete RRT for Coordinated Multi-robot Planning</a>. Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - (Volume 2).</p> <p><em>@inproceedings{Hvezda18icinco,<br> author = {Hvězda, Jakub and Kulich, Miroslav and Přeučil, Libor},<br> title = {Improved Discrete RRT for Coordinated Multi-robot Planning},<br> booktitle = {Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics - (Volume 2)},<br> publisher = {SciTePress},<br> address = {Madeira, PT},<br> year = {2018},<br> language = {English},<br> url = {http://www.scitepress.org/PublicationsDetail.aspx?ID=ppwUqsGaX18=\&t=1},<br> doi = {10.5220/0006865901710179},<br> access = {full}<br> }</em></p> <p> </p> <p> </p> <p> </p>
renardbebe/Smart-Contract-Benchmark-Suites: Smart-Contract-Benchmark-Suites v0.4.0
<p>A unified smart contract standard data set.</p> <p>Contains: dataset and used tools in paper "<em><strong>Empirical Evaluation of Smart Contract Testing: What Is the Best Choice? </strong></em>"</p>
EvoBench: Benchmarking Schema Evolution in NoSQL
<p>Docker containers for reproducing the proof of concept measurements with our NoSQL Schema Evolution Benchmark.</p>
Data for: A Benchmark Engineering Methodology to Measure the Overhead of Application-Level Monitoring
<p>Application-level monitoring frameworks, such as Kieker, provide insight into the inner workings and the dynamic behavior of software systems. However, depending on the number of monitoring probes used, these frameworks may introduce significant runtime overhead. Consequently, planning the instrumentation of continuously operating software systems requires detailed knowledge of the performance impact of each monitoring probe.</p> <p>In this paper, we present our benchmark engineering approach to quantify the monitoring overhead caused by each probe under controlled and repeatable conditions. Our developed MooBench benchmark provides a basis for performance evaluations and comparisons of application-level monitoring frameworks. To evaluate its capabilities, we employ our benchmark to conduct a performance comparison of all available Kieker releases from version 0.91 to the current release 1.8.</p> <p>This dataset supplements the paper and contains the raw experimental data as well as several generated diagrams for each experiment.</p>
Benchmark Database for Phonetic Alignments
<p>In the last two decades, alignment analyses have become an important technique in quantitative historical linguistics and dialectology. Phonetic alignment plays a crucial role in the identification of regular sound correspondences and deeper genealogical relations between and within languages and language families. Surprisingly, up to today, there are no easily accessible benchmark data sets for phonetic alignment analyses. Here we present a publicly available database of manually edited phonetic alignments which can serve as a platform for testing and improving the performance of automatic alignment algorithms. The database consists of a great variety of alignments drawn from a large number of different sources. The data is arranged in a such way that typical problems encountered in phonetic alignment analyses (metathesis, diversity of phonetic sequences) are represented and can be directly tested.</p>
Data for: Performance Benchmarking of Application Monitoring Frameworks
<p>Application-level monitoring of continuously operating software systems provides insights into their dynamic behavior, helping to maintain their performance and availability during runtime. Such monitoring may cause a significant runtime overhead to the monitored system, depending on the number and location of used instrumentation probes. In order to improve a system’s instrumentation and to reduce the caused monitoring overhead, it is necessary to know the performance impact of each probe.<br /> While many monitoring frameworks are claiming to have minimal impact on the performance, these claims are often not backed up with a detailed performance evaluation determining the actual cost of monitoring. Benchmarks can be used as an effective and affordable way for these evaluations. However, no benchmark specifically targeting the overhead of monitoring itself exists. Furthermore, no established benchmark engineering methodology exists that provides guidelines for the design, execution, and analysis of benchmarks.<br /> This thesis introduces a benchmark approach to measure the performance overhead of application-level monitoring frameworks. The core contributions of this approach are 1) a definition of common causes of monitoring overhead, 2) a general benchmark engineering methodology, 3) the MooBench micro-benchmark to measure and quantify causes of monitoring overhead, and 4) detailed performance evaluations of three different application-level monitoring frameworks. Extensive experiments demonstrate the feasibility and practicality of the approach and validate the benchmark results. The developed benchmark is available as open source software and the results of all experiments are available for download to facilitate further validation and replication of the results.</p> <p>This dataset supplements the thesis and contains the results of all experiments, including the raw result data, the results of additional experiments, and the configuration of our benchmarks.</p>
Benchmark Hypergraphs and Detailed Experimental Results of "k-way Hypergraph Partitioning via n-Level Recursive Bisection"
<p>This dataset contains hypergraphs derived from three benchmark sets: The<br /> ISPD98 VLSI Circuit Benchmark Suite [1], the University of Florida Sparse Matrix Collec-<br /> tion [2] and the international SAT Competition 2014 [3]. From the latter, we randomly selected<br /> 100 instances from the application track and converted them into hypergraphs as follows:<br /> Each boolean variable (and its complement) is mapped to one vertex and each clause constitutes<br /> a net [41]. The Sparse Matrix Collection is organized into 172 groups and each group contains<br /> matrices of different application areas. From each group, we choose one matrix for each appli-<br /> cation area that has between 10 000 and 10 000 000 columns. In case multiple matrices fulfill<br /> our criteria, we randomly select one. In total, we include 192 matrices, which are translated into<br /> hypergraphs using the row-net model, i.e. each row is treated as a net and each column as<br /> a vertex. Empty rows are discarded. Both vertices and nets have unit weight. Together with the<br /> 18 ISPD98 VLSI instances , a total of 310 hypergraphs constitute our benchmark set. 4 Each of<br /> these hypergraphs is partitioned into k ∈ {2, 4, 8, 16, 32, 64, 128} blocks with ε = 0.03. For each<br /> value of k, a k-way partition is considered to be one test instance, resulting in a total of 2170<br /> instances.</p> <p>See the README for further information on the different files contained in this dataset.</p> <p>[1 ]C. J. Alpert. The ISPD98 Circuit Benchmark Suite. In Proc. of the 1998 Int. Symp. on Physical Design, ISPD ’98, pages 80–85, New York, 1998. ACM.<br /> [2] T. A. Davis and Y. Hu. The University of Florida Sparse Matrix Collection. ACM Trans. Math. Softw.,38(1):1:1–1:25, 2011.<br /> [3] A. Belov, D. Diepold, M. Heule, and M. Järvisalo. The SAT Competition 2014. http://www.satcompetition.org/2014/, 2014.</p>
graphANNIS benchmark queries and results
<p>Queries used in the benchmark are included in the file "queries_graphannis_benchmark.zip". Each folder corresponds to one corpus. Each query has an ".aql" file containing the AQL query, a ".json" file with the parsed query.</p> <p>The actual results of the benchmark are in the "journal-2016-07-27.csv " file. Additional the memory usage for each corpus in graphANNIS is given in "memoryusage.csv".</p>
Simulated benchmark metagenome used to demonstrate and evaluate MGLEX software
<p>This is a mock dataset of 120 000 artificial contigs of 1 kb length derived by simulating reads from 295 unique genomes and 44 species with each two or three strain genomes using the ART read simulator (Huang et al., 2012) and a lognormal abundance distribution. Genomes were chosen according to the CAMI2015 (www.cami-challenge.org) medium complexity toy dataset. The dataset contains four replicate samples with varied abundances and corresponding sequence feature files in MGLEX v0.1.1 format to use for genome reconstruction. Our aim was to create a benchmark dataset under controlled settings, minimizing potential biases introduced by specific software. This package also includes MGLEX benchmark scripts.</p>
Use Case Points Benchmark Dataset
<p>This dataset was gathered by us from three software houses. This is real-life dataset. Use Case points method as originated by Karner was used for counting a steps or number of actors. Data are based on different languages, various problem domains. ISBSG style for language, domain and application type were adopted.</p> <p><br> Attributes are used as follows:<br> Project_No - only project ID for identification purposes<br> Simple Actors - Number of actor classify according UCP - simple actors.<br> Average Actors - Number of actor classify according UCP - average actors.<br> Complex Actors - Number of actor classify according UCP - complex actors.<br> UAW - Unadjusted Actor weight, computed by using UCP equation. <br> Simple UC - Number of use cases classified as simple - UCP number of steps is used.<br> Average UC - Number of use cases classified as average - UCP number of steps is used.<br> Complex UC - Number of use cases classified as complex - UCP number of steps is used.<br> UUCW - Unadjusted UseCase Weight - computed by using UCP equation.<br> TCF - Technical Complexity FactorECF - Enviromental Complexity Factors<br> Real_P20 - Real_P20 - Real Effort in Person hours, decided by productivity factor (PF = 20).<br> Real_Effort_Person_Hours - Real Effort (development time) in person-hours.<br> Sector - Problem domain of projectLanguage - Programming language used for project.<br> Methodology - Development methodology used for project development.<br> ApplicationType - Classification of project type - provided by donator. <br> DataDonator - anonymized acronym for data donator.<br> </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.