Find research datasets worth reusing

opencc-by-4.0Jul 2019View details →

Consistency and differences between centrality measures across distinct classes of networks

This is the data reported in Oldham et al., 2019. Consistency and differences between centrality measures across distinct classes of networks (<a href="https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0220061">link to paper</a>). Originally hosted <a href="https://figshare.com/s/22c5b72b574351d03edf">here</a>. GitHub for the code used to generate and use this data is located <a href="https://github.com/StuartJO/CentralityConsistency">here</a>. Unzip the zip file into the main directory into a folder called "Data".

opencc-by-4.0Dec 2023View details →

Dataset and codes: Abundance of trace fossil Phycosiphon incertum in core sections measured using a convolutional neural network

Open the record for dataset details and reuse information.

opencc-by-4.0May 2021View details →

Data for paper: Comparing Measures of Centrality in Bipartite Social Networks: A Study of Drug Seeking for Opioid Analgesics

Data and code for paper: Comparing Measures of Centrality in Bipartite Social Networks: A Study of Drug Seeking for Opioid Analgesics

opencc-by-4.0Aug 2024View details →

Measuring Network Latency from a Wireless ISP: Variations Within and Across Subnets - Dataset

<h1>Overview</h1> This archive contains the data from a network latency measurement campaign conducted inside the wireless ISP JackRabbit Wireless between 2023-10-30 and 2023-12-01. The measurement setup and an analysis of the data is provided in the paper "Measuring Network Latency from a Wireless ISP: Variations Within and Across Subnets", accepted for the ACM Internet Measurement Conference 2024.  Code for parsing the data and reproducing the plots can be found at <a href="https://github.com/simosund/Measuring-Network-Latency-from-a-Wireless-ISP.git">https://github.com/simosund/Measuring-Network-Latency-from-a-Wireless-ISP.git</a> <h2>Manifest</h2> <ul> <li>The raw subfolder: contains the original data captured by epping from the measurement campaign. However, note that we have merged all subnet entries from the internal network into a single entry at each timepoint to avoid revealing details about the subnet structure on the internal network. Some internal subnet entries that only carry control traffic have also been removed. It is still possible to recreate all the results from the paper with this data.</li> <li>The processed subfolder: This contains processed versions of the data that have been merged in a few different ways, allowing the results from the paper to be recreated quickly without having to parse all of the original data. <ul> <li>subnets_stats_merged.json: The data from the entire measurement period have been merged into a single entry per subnet</li> <li>{lan,wan}_per10s_stats_merged.json: All the internal or external data merged into a single entry per timepoint (10s).</li> <li>top_100asn_stats_10min_json: The external data merged per ASN (for the 100 ASNs with the highest amount of downlink traffic during the measurement period) at a 10 minute resolution.</li> <li>protocol_counters.hdf5: Counters for number of packets of each <a href="https://www.iana.org/assignments/protocol-numbers/protocol-numbers.xhtml">IP-protocol</a>. The most common protocols, such as TCP, UDP and ICMP are named, where less common ones are just numbered.</li> <li>ecn_counters.hdf5: Counters of the four ECN-marks (Not-ECT, ECT(1), ECT(0) and CE).</li> </ul> </li> <li>The overhead_measurements subfolder: CPU and network load measurements from a later period used to estimate the overhead of the epping tool. In the covered period, epping was running between the 19th of April and the 3rd of May. <ul> <li>InfluxDB_CPU_20240413-20240513.csv: CSV with hourly CPU utilization per core on the DuT.</li> <li>jackrabbit-shaper-per\_day-20240421-20240513.csv: CSV with network load metrics (throughput in bytes and packets) every 48 minutes.</li> <li>proglistoutput_after_enabling_bpf_stats_24hr.txt: Output from <code>bpftool prog list</code> after enabling <code>sysctl -w kernel.bpf_stats_enabled=1</code>, containing run counts and cumulative runtime for all eBPF programs on the system.</li> </ul> </li> <li>rrc14_20231115_ipasn.dat: Data from RIPE´s route collector RRC14 in the middle of the measurement campaign (2023-11-15), used to map subnets into ASNs.</li> <li>asn.txt: RIPE's list of AS names (from <a href="https://ftp.ripe.net/ripe/asnames/asn.txt">https://ftp.ripe.net/ripe/asnames/asn.txt</a>), used to translate ASNs into AS names.</li> </ul> <h2>Data format for raw data</h2> The raw data is organized into one subfolder per day, with each subfolder containing one compressed json-file per minute. All files have been included in a tar-archive. The JSON format in each file consists of a single list of entries. There are three types of entries. <h3>Configuration entry</h3> Each file starts with one configuration entry, containing the following fields: <ul> <li> timestamp: Unix timestamp in nanoseconds when entry was created</li> <li>bins: The maximum number of bins in the histogram of a subnet entry. Always 250 in this dataset. Note that histograms may contain less bins as unused bins in the upper range are truncated.</li> <li>bin_width_ns: The width of the bins in nanoseconds. Always 4000000 (4ms) in this dataset.</li> <li>aggregation_interval_ns: How often the aggregated stats are reported in nanoseconds. Always 10000000000 (10s) in this dataset.</li> <li>timeout_interval_ns: How long epping will keep a subnet entry around since it last saw traffic for it. Always 30000000000 (30s) in this dataset. Note that a subnet entry may only be deleted after its stats are reported, so no stats will be lost regardless of value.</li> <li>ipv4_prefix_len: Subnet size to aggregate IPv4 traffic. Always 24 (/24) in this dataset.</li> <li>ipv6_prefix_len: Subnet size to aggregate IPv6 traffic. Always /48 in this dataset.</li> </ul> Example: <pre><code>{"timestamp": 1698661062033443066, "bins": 250, "bin_width_ns": 4000000, "aggregation_interval_ns": 10000000000, "timeout_interval_ns": 30000000000, "ipv4_prefix_len": 24, "ipv6_prefix_len": 48}</code></pre> <h3> Subnet stats entry</h3> The most common entry will be the subnet stats entry, which contains the aggregated stats for one /24 subnet (or /48 if IPv6) over a 10 second period. Each entry always contains the following fields: <ul> <li>timestamp: A unix timestamp in nanoseconds of when the entry were reported. The entry will thus contain the stats for the period [timestamp - 10s, timestamp].</li> <li>ip_prefix: The subnet in standard IPv4 or IPv6 CIDR notation.</li> <li>rx_stats and tx_stats: Packet and byte counts for packets received from/transmitted to the subnet. Note that rx and tx is from the perspective of the epping node, not the perspective of the subnet, so rx is packets epping has seen sent from the subnet, and tx is packets epping has seen sent towards the subnet. <ul> <li>TCP_TS, TCP_noTS and other: Entries for TCP traffic that has TCP timestamps enabled, TCP traffic with TCP timestamps disabled, and non-TCP traffic, respectively. The entry will be omitted if there is no traffic of the relevant type. <ul> <li>packets: Number of packets of the specific type</li> <li>bytes: Sum of bytes in the observed packets, including headers (up to and including the 14-byte Ethernet header).</li> </ul> </li> </ul> </li> </ul> Additionally, the following fields will be present if any RTT has been observed for the subnet. Note that the subnet will include RTTs for traffic going towards the subnet, i.e. SRC -> subnet -> SRC. <ul> <li>min_rtt: The minimum observed RTT in nanoseconds.</li> <li>max_rtt: The maximum observed RTT in nanoseconds.</li> <li>histogram:: A histogram with all observed RTTs. The bin width and maximum number of bins is specified by the configuration entry (always 4ms wide bins and max 250 bins in this dataset). Note that the histogram will truncate 0-bins after the last non-zero bin (e.g. if the maximum RTT is 7 ms, the histogram will only contain 2 bins).</li> <li>count_rtt: The total number of RTTs observed (sum of histogram bin counts)</li> <li>mean_rtt: The approximate mean RTT in nanoseconds.</li> <li>median_rtt: The approximate median RTT in nanoseconds.</li> <li>p95_rtt: The approximate 95th percentile of the RTTs in nanoseconds</li> </ul> Note that mean_rtt, median_rtt and p95_rtt are all calculated from the histogram, assuming all RTTs are in the middle of the histogram bins, and will thus be +/- 2ms of the real value in this dataset. Example: <div> <pre><code>{"timestamp": 1698661072800129847, "ip_prefix": "100.64.0.0/24", "rx_stats": {"TCP_TS": {"packets": 64816, "bytes": 23063341}, "other": {"packets": 39783, "bytes": 16074723}, "TCP_noTS": {"packets": 8505, "bytes": 7403649}}, "tx_stats": {"TCP_TS": {"packets": 134424, "bytes": 175370347}, "other": {"packets": 76267, "bytes": 86104300}, "TCP_noTS": {"packets": 7946, "bytes": 5174210}}, "min_rtt": 50766, "max_rtt": 440315940, "histogram": [622, 596, 835, 1106, 932, 410, 228, 139, 108, 68, 78, 79, 73, 63, 49, 35, 22, 33, 31, 24, 21, 18, 23, 15, 20, 20, 37, 35, 96, 113, 95, 80, 36, 22, 24, 27, 17, 31, 9, 9, 14, 18, 10, 21, 22, 21, 22, 25, 8, 9, 16, 8, 12, 9, 7, 5, 6, 6, 3, 4, 4, 2, 2, 1, 1, 3, 3, 1, 0, 3, 2, 0, 1, 1, 1, 1, 2, 2, 1, 2, 1, 0, 0, 0, 1, 2, 0, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1], "count_rtt": 6568, "mean_rtt": 37529232.64311815, "median_rtt": 18000000.0, "p95_rtt": 150000000.0}</code></pre> </div> <div> <h3>Global counters entry</h3> Each report interval will also contain one entry with global counters, with the following fields: <ul> <li>timestamp: Unix timestamp in nanoseconds of when the entry was reported. The entry will thus contain stats for the period [timestamp - 10s, timestamp].</li> <li>protocol_counters: A set of fields with packet and byte counters for the <a href="https://www.iana.org/assignments/protocol-numbers/protocol-numbers.xhtml" rel="nofollow">protocols indicated in the IP-header</a>. The most common protocols will be named, specifically TCP, UDP, ICMP, ICMPv6 and the special non-IP for non-IP packets. All other protocols will simply use the protocol number. Protocols for which no traffic has been observed are omitted. <ul> <li>packets: The number of packets of the specific type</li> <li>bytes: Sum of bytes in the observed packets, including headers (up to and including the 14-byte Ethernet header). Only the list of named protocols above will include this byte counter.</li> </ul> </li> <li>ecn_counters: Counters for the four distinct ECN code points in the IP header. May contain the fields no_ECT (00), ECT1 (01), ECT0 (10) and CE (11).</li> </ul> </div> Example: <div> <pre><code>{"timestamp": 1698661072848798575, "protocol_counters": {"TCP": {"packets": 249032, "bytes": 242282230}, "UDP": {"packets": 119778, "bytes": 108251262}, "ICMP": {"packets": 1523, "bytes": 445347}, "ICMPv6": {"packets": 87, "bytes": 25031}, "41": {"packets": 1}, "47": {"packets": 251}, "89": {"packets": 22}}, "ecn_counters": {"no_ECT": 347980, "ECT1": 2, "ECT0": 22709, "CE": 3}, "errors": {}}</code></pre> </div>

ClinicalTrials.gov32/100

Telemedical, Intersectoral Network as New Digital Health Structure to Measurably Improve the Local Health Care

ClinicalTrials.gov study NCT03137589. IPD Sharing: NO. Countries: 1. Publications: 4.

closedIPD-NOFeb 2026View details →

dryad32/100

Data from: Organic functional group and organic matter concentrations from FT-IR measurements of particulate matter samples in the Southeastern Aerosol Research and Characterization (SEARCH) network from 2009-2016

Open the record for dataset details and reuse information.

publicNov 2021View details →

dryad32/100

Data from: Node-based measures of connectivity in genetic networks

Open the record for dataset details and reuse information.

publicApr 2015View details →

dryad32/100

Dataset for: Physics-informed neural networks with monotonicity constraints for Richardson-Richards equation: Estimation of constitutive relationships and soil water flux density from volumetric water content measurements by Toshiyuki Bandai and Teamrat A. Ghezzehei

Open the record for dataset details and reuse information.

publicMar 2022View details →

opencc-by-4.0Jul 2020View details →

Developing a deep Learning network to retrieve ocean hydrographic profiles in the North Atlantic from combined satellite and in situ measurements: training datasets

We provide here the datasets used for the development of a deep learning algorithm which is presently candidate for the development of a daily 3D ocean product covering the North Atlantic at 1/10° resolution, over the 2010-2018 period, as part of the European Space Agency World Ocean Circulation project (ESA-WOC). The method is based on a stacked Long Short-Term Memory neural network, coupled to a Monte-Carlo dropout approach, and allows to project satellite-derived sea surface temperature, sea surface salinity and absolute dynamic topography data at depth after training with sparse co-located in situ vertical hydrographic profiles (Buongiorno Nardelli, 2020, doi:<a href="https://www.researchgate.net/deref/http%3A%2F%2Fdx.doi.org%2F10.3390%2Frs12193151?_sg%5B0%5D=0xE-347r7Hvb80klJcEo811AhUiXq-twG_E6l4yB-BfIKkVtW-lVLGcO02mTFkUczvozYYI0WCPyUBFR3kzWNGGZKg.ftvLheFrzHIJriO4qW2bdxalvR_TWt3MpwUfvto3EemhRgvDRGwJ9Mdy4Xr0IcGCfICivf4j-VqTgKxVvXRogA">10.3390/rs12193151</a>).  The training/test dataset presented here includes different sets of co-located temperature and salinity vertical profiles and corresponding satellite surface data:  <ul> <li>in situ observations extracted from the quality controlled Argo and CTD profiles produced by Copernicus Marine Environment Monitoring Service CORA 5.2 (<a href="http://marine.copernicus.eu/services-portfolio/access-to-products/">http://marine.copernicus.eu/services-portfolio/access-to-products/</a>, product_id: INSITU_GLO_TS_REP_OBSERVATIONS_013_001_b, doi: 10.17882/46219TS1, Szekely et al., 2019) and interpolated through a spline on a regularly spaced vertical grid (with 10 m intervals);</li> <li>climatological profiles extracted from World Ocean Atlas 2013 optimally interpolated monthly fields (Locarnini et al., 2013; Zweng et al., 2013), interpolated through a spline on a regularly spaced vertical grid (with 10 m intervals), upsized to a 1/10° horizontal grid through a cubic spline and linearly interpolated in time between the central day of each month;</li> <li>co-located Sea Surface Temperature taken from the level 4 (L4, i.e. interpolated) multi-year reprocessed Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA) developed by U.K. Met Office and distributed (upon free registration) through the Copernicus Marine Environment Monitoring Service (CMEMS, <a href="http://marine.copernicus.eu/services-portfolio/access-to-products/">http://marine.copernicus.eu/services-portfolio/access-to-products/</a>, product_id=SST_GLO_SST_L4_REP_OBSERVATIONS_010_011);</li> <li>co-located Sea Surface Salinity taken from dataset developed within ESA-WOC project (https://doi.org/10.5281/zenodo.3943813);</li> <li>co-located Absolute Dynamic Topography (ADT) data distributed by CMEMS as reprocessed data (<a href="http://marine.copernicus.eu/services-portfolio/access-to-products/">http://marine.copernicus.eu/services-portfolio/access-to-products/</a>, product_id: SEALEVEL_GLO_PHY_L4_REP_OBSERVATIONS_008_047), upsized here to the ESA-WOC 1/10°x1/10° grid through a cubic spline and adjusted to insitu steric heights by regressing steric heights and co-located ADT data in the neighbourhood of each grid point, considering matchups within a temporal window of 10 days (as in Buongiorno Nardelli et al., 2017).</li> </ul>

opencc-zeroDec 2012View details →

Data from: Accurate measurements of dynamics and reproducibility in small genetic networks

Quantification of gene expression has become a central tool for understanding genetic networks. In many systems the only viable way to measure protein levels is by immunofluorescence, which is notorious for its limited accuracy. Using the early Drosophila embryo as an example, we show that careful identification and control of experimental error allows for highly accurate gene expression measurements. We generated antibodies in different host species, allowing for simultaneous staining of four Drosophila gap genes in individual embryos. Careful error analysis of hundreds of expression profiles reveals that less than ∼20% of the observed embryo-to-embryo fluctuations stem from experimental error. These measurements make it possible to extract not only very accurate mean gene expression profiles but also their naturally occurring fluctuations of biological origin and corresponding cross-correlations. We use this analysis to extract gap gene profile dynamics with ∼1 min accuracy. The combination of these new measurements and analysis techniques reveals a two-fold increase in profile reproducibility due to a collective network dynamics that relays positional accuracy from the maternal gradients to the pair-rule genes.

opencc-zeroDec 2015View details →

Data from: Can longitudinal generalized estimating equation models distinguish network influence and homophily? an agent-based modeling approach to measurement characteristics

Background: Connected individuals (or nodes) in a network are more likely to be similar than two randomly selected nodes due to homophily and/or network influence. Distinguishing between these two influences is an important goal in network analysis, and generalized estimating equation (GEE) analyses of longitudinal dyadic network data are an attractive approach. It is not known to what extent such regressions can accurately extract underlying data generating processes. Therefore our primary objective is to determine to what extent, and under what conditions, does the GEE-approach recreate the actual dynamics in an agent-based model. Methods: We generated simulated cohorts with pre-specified network characteristics and attachments in both static and dynamic networks, and we varied the presence of homophily and network influence. We then used statistical regression and examined the GEE model performance in each cohort to determine whether the model was able to detect the presence of homophily and network influence. Results: In cohorts with both static and dynamic networks, we find that the GEE models have excellent sensitivity and reasonable specificity for determining the presence or absence of network influence, but little ability to distinguish whether or not homophily is present. Conclusions: The GEE models are a valuable tool to examine for the presence of network influence in longitudinal data, but are quite limited with respect to homophily.

openother-openOct 2023View details →

Replication package for "Measuring Diffusion over a Large Network"

The package contains data sets and codes to replicate the tables and figures in He and Song (forthcoming), "Measuring Diffusion Over a Large Network" Review of Economic Studies. Detailed instructions to obtain some of the raw data are given in the package.

opencc-by-4.0Dec 2023View details →

Simulation and experimental data of frequency domain and time domain optical signal measurements for optical network digital twins

The dataset contains IQ optical constellation samples for 16-QAM optical connections. Data have been generated both experimentally and through simulations with a MATLAB-based simulator. Different configurations have been simulated: 62 lightpaths having a different number of spans and links and 4 soft-failures affecting a lightpath with increasing failure magnitude.

opencc-by-4.0Nov 2016View details →

Supplementary material 1 from: Sprondel N, Donner J, Mahlkow N, Köppel J (2016) Urban climate and heat stress: how likely is the implementation of adaptation measures in mid-latitude cities? The case of façade greening analyzed with Bayesian networks. One Ecosystem 1: e9280. https://doi.org/10.3897/oneeco.1.e9280

Questionnaire for Bayesian network analysis

restrictedcc-by-4.0Oct 2024View details →

DATA_measure and predict the travel time reliability on the urban rail transit network

(1)The automatic fare collection data recorded the time of entering and exiting stations for each trip. (2)The automatic fare collection data format is CSV, and the columns of the table are in Chinese. The columns are ticket card number, ticket card type, date, time of entering the station, entering station name, time of exiting the station, exiting station name, and day of the week.

publicApr 2020View details →

Data from: Accurate measurements of dynamics and reproducibility in small genetic networks

Open the record for dataset details and reuse information.