201724 results
Contributors: Diane Rekow, Jean-Yves Baudouin, Bruno Rossion, Arnaud Leleu
Date: 2020-02-14
... Preprocessed EEG data, behavioral measures, participants infos and experiment code for the study 'An ecological measure of rapid and automatic face-sex categorization'. Please find additional information in 'READ_ME.txt' file.
Data types:
Contributors: Gábor Szűcs, Dávid Papp
Date: 2020-02-14
... In active learning, Optimally Balanced Entropy-Based Sampling (OBEBS) method is a selection strategy from unlabelled data. At active zero-shot learning there is not enough information for supervised machine learning method, thus, our sampling strategy was based on unsupervised learning (clustering). The cluster membership likelihoods of the items were essential for the algorithm to connect the clusters and the classes; i.e. to find assignment between them. For best assignment, Hungarian algorithm was used. We developed and implemented adaptive assignment variants of OBEBS method in the software.
Data types:
Contributors: Gleb Dyakonov
Date: 2020-02-14
... The dataset contains information on the currents in the upper 15-meter layer of the Caspian Sea in 2003-2005. The velocity fields were reconstructed in an eddy-resolving ocean general circulation model SZ-COMPAS using a realistic forcing. The data arrays have a very high resolution: ~2 km in space and 4 hours in time. This is sufficient to resolve most of the mesoscale features of sea dynamics as well as a wide range of their temporal spectrum, including inertial oscillations, synoptic and seasonal variability. The NetCDF files are: 3 files with instantaneous currents and 3 files with monthly mean currents (suffix “mm” in the file names). Each file corresponds to one of the three horizons (depths): 1 m, 7 m, and 15 m. In the horizontal plane the data are defined on a uniform geographical grid (46.7625–54.2125°E, 36.5092–47.2892°N), dimensions of all of the arrays are 299 by 589. The period covered by the instantaneous data is 2003, time step is 4 hours; the first record corresponds to 2003-01-01 04:00:00 GMT. The period covered by the monthly mean data is 2003-2005 with one vector field for every month, defined on the last day of the month. All data dimension is cm/s. It should be noted that the model describes the upper 30-meter layer of the sea in sigma-coordinate, so the data values are actually defined on the 1st, 4th, and 8th sigma-horizons, rather than 1 m, 7 m, and 15 m depths. This means that, in the sea cells with bottom depth less than 30 m, the actual depth of the data nodes is ~3%, ~23% and 50% of real water column height, while in the rest of cells (with bottom depth greater than 30 m), the data nodes are located at ~1 m, ~7 m, and ~15 m of depth (± few centimeters depending on the local sea level).
Data types:
Top results from Data Repository sources. Show only results like these.
Contributors: Ricardo R Palma
Date: 2020-02-13
... Main authors potentially importatn from the chimical epistemic field in byopolimers. (from 2006 to 2019) - Patents registered and papers (Conferences are not included. Database in .bib bibliographic format.
Data types:
Contributors: Beatrice Gordon, Ginger Paige, Scott Miller, Niels Claes, Andrew Parsekian
Date: 2020-02-12
... Attached are data that were used to calculate return flow for this study. We do not include geophysical data as it is too large and is still a niche field where few people are likely excited to work with those data. We will revisit publication of geophysical data if it turns out there is interest in using those data.
Data types:
Contributors: Elena Dumitrescu, Sullivan Hué, Christophe Hurlin, Sessi Tokpavi
Date: 2020-02-11
... Datasets and codes of the article : Machine Learning or Econometrics for Credit Scoring: Let's Get the Best of Both Worlds
Data types:
Contributors: Milad Mohammadalizadehkorde
Date: 2020-02-11
... Raw tweets of 4 Fridays in August 2017. This data has been gathered through Twitter API which covers a sample of all tweets available on Fridays 4, 11, 18 and 25 August 2017.
Data types:
Contributors: Daniela Jovanovska
Date: 2020-02-11
... Providing information on the spatial distribution of habitat groups through land cover classification is useful for a wide range of environmental studies, especially those monitoring and managing succession and land cover change. While Corine Land Cover is widely used in standardised land cover comparisons, its strength is primarily its use in coarse scale analyses over large extent. Recently available Sentinel-2 combined with wide availability of open-source remote sensing software allows customised fine-scale land cover classifications to suit research projects applied over a smaller extent. In this regard, this study aims to test the suitability of Sentinel-2 imagery in providing a customised land cover classification for Shar Planina mountain range in North Macedonia, hoping that the results will serve as a background for assessment of its landscape visual quality as part of the ongoing PhD study (Jovanovska, D. 2018-2021). The results show that use of Sentinel-2 multiband high resolution images allows customized medium-large scale image classification that yields great accuracy results. Furthermore, frequent updating of Sentinel-2 database allows comparability and standardised land cover change analysis for the purpose of environmental monitoring. The resulting land cover data classification is uploaded as an .ASCII file accompanied with corresponding .lyr file and can serve as a background for subsequent environmental studies in the region. Finally, any further improvements of the final output, as well as trials on its suitability in studies with different aim and scope are encouraged and welcomed.
Data types:
Contributors: Carme Uribe
Date: 2020-02-11
... dataset including results of the Data in Brief "Data for functional MRI connectivity in transgender people with gender incongruence and cisgender individuals" supplementary data from the "Brain network interactions in transgender individuals with gender incongruence" manuscript submitted to Neuroimage journal Original T1- and T2*-weighted images can be found in: doi:10.17632/hjmfrv6vmg.1
Data types:
Contributors: George Busby
Date: 2020-02-10
... Here is a combined dataset of genetic data on 2,643 individuals from 163 worldwide human populations. These genotypes were all generated on Illumina chips (550, 610, 660) for multiple different studies. The two main papers that this dataset was compiled for are: Hellenthal, et al 2014 A Genetic Atlas of Human Admixture History, Science; and Busby, et al 2015 The role of recent admixture in forming the contemporary West Eurasian genomic landscape, Current Biology. The data are in PLINK format and the BusbyWorldwidePopulations.csv file outlines where the different datasets come from. Note that because these two datasets were combined together, not all populations are typed on the same set of SNPs. We have included genotype data on 523,443 SNPs, of which 441,038 are genotyped on at least 97.5% of individuals. Therefore, additional QC steps are required to filter this set down to high quality calls, depending on the subset of samples that are required. Complete information about the populations used is available in the various publications that are outlined in the associated paper. Note that these same populations are available elsewhere and this dataset represents that compiled for the above mentioned papers. UPDATE 11/11/2019 Thanks to some heroic work by Kristján Helgi Swerford Moore at DECODE, I have now updated the population and sample information to more accurately and verbosely label the individuals.
Data types: