The Department of Housing, Local Government and Heritage publishes annual RAPTOR (Recording and Addressing Persecution and Threats to Our Raptors) reports on threats to birds of prey. This csv dataset for download here represents the tabular data that is core to those reports. It provides details of recorded incidents of human related injury and mortality in Irish birds of prey, as well as any incidents of poisoned bait or poisoning of any wildlife. This dataset should be viewed in conjunction with its associated 2015 report which is also referenced here for download. The dataset and report is prepared by the National Parks and Wildlife Service (NPWS) in collaboration with the Regional Veterinary Labs of the Department of Agriculture, Food and the Marine, and the State Laboratory. The report is the product of a joint Departmental initiative to investigate bird of prey deaths in Ireland. The dataset enables an appraisal of black spots, associated land-use types, methods of persecution, motives behind the persecution and the times of year at which such incidents peak. 2015 saw the largest annual number of incidents since recording began systematically in 2011. In total, 35 poison and persecution incidents were confirmed. Poisoning falls into two general categories: accidental poisoning through the use of poison against rats and mice which then accumulates in birds that eat them, most notably red kites and barn owls; and deliberate laying of poison. The victims of poisoning and persecution since 2007 include Red Kite, Common Buzzard, Peregrine Falcon, Golden Eagle, White-tailed Sea Eagle, Sparrowhawk, Kestrel, Hen Harrier, Barn Owl and Short-eared Owl. More than a hundred other birds such as crows and pigeons were also found to have been poisoned. .hidden { display: none }

This data set describes genomic sequence information from 2022 used to infer spatiotemporal trends pertaining to the introductions of highly pathogenic H5N1 avian influenza viruses into Alaska and spread among wild birds, backyard poultry, and mammals.

Historic bird observations by R. Paefsler on 5 voyages from Europe to Pacific Ocean from 1910 to 1913 digitised from cited publications (see below). The records are principality 'presence' only with only a few with observed counts. The recorded scientific names were matched to current names where possible using the Biodiversity Heritage Library, web searches and noting some names were slightly misspelt. There are a few names that could not be matched and therefore are not linked via scientificnameId to WoRMS. Place names typically ports or major towns were georeferenced and recorded to within 2000 metres of the likely location. Original publications used are Paefsler, R. Beiträge zur Verbreitung der Seevögel. J. Ornithol 62, 272–278 (1914). https://doi.org/10.1007/BF02096342 and Paefsler, R. Beiträge zur Verbreitung der Seevögel. J. Ornithol 61, 41–51 (1913). https://doi.org/10.1007/BF02250435 Digtised data (as CSV) is available at https://www.marine.csiro.au/data/trawler/download.cfm?file_id=5539

These data support a paired USGS publication and document the use of retention ponds on commercial poultry farms by wild waterfowl.

Understanding the relationship between biodiversity and ecological stability is increasingly urgent as rapid species extinction continues. Though evidence of positive diversity-stability relationships is accumulating, empirical results are inconsistent and effect sizes tend to be small, raising questions about relative contributions of intrinsic (i.e., species composition/interactions) and extrinsic (i.e., environmental) drivers of stability. Community stability may be more strongly influenced by environmental conditions than by community diversity in some contexts, yet little is known about the comparative importance of diversity and climate means, extremes, and variability in regulating stability. We used a half-century of continental-scale bird data to quantify avian community temporal variability (a metric often used to approximate ecological stability) at 1,379 sites and compared relative effects of climatic variables and species richness. We found that extreme heat and extremely l..., , , # Code from: The relative influence of climate extremes and species richness on the temporal variability of bird communities

Date: Mar 30, 2023

Lead author contact: Samantha M. Cady, University of Nebraska-Lincoln (samantha.cady@unl.edu)

Additional authors/cooperators: Samuel D. Fuhlendorf (Oklahoma State University), Craig A. Davis (Oklahoma State University), Barney Luttbeg (Oklahoma State University), Caleb P. Roberts (University of Arkansas), and Scott R. Loss (Oklahoma State University)

DATA

User must download 5 open-source datasets and 1 dataset from Cady et al. (2023) on Dryad before running the accompanying R Code.

Open Source Data:

Datasets 1-3: Breeding Bird Survey 2019 Release, state-level data 1966-2018, weather.csv, and coords.csv. All three datasets are available online:

Pardieck, K.L., Ziolkowski, D.J., Lutmerding, M., Aponte, V.I. & Hudson, M.-A.R. (2019). North American Breeding Bird Survey Dataset 1966-2018 (ver. 2018.0). ...

This dataset contains seabird survey and associated environmental conditions data collected in the Chukchi Sea during the 5 August - 25 August 2017 Arctic Marine Biodiversity Observing Network (AMBON) research cruises. Seabirds were surveyed using line transect methods. Seabird observations were recorded directly into a laptop computer using software which logged the geographic coordinates of each sighting. Incidental sightings of marine mammal were also recorded. Supplementary environmental data are included for the beginning of each survey period. The dataset is two comma-separated values (csv) files. The file named dlog_2017_dwc_core.csv contains the count of observed marine birds and mammals by species, their behavior, and the environmental conditions recorded during the surveys for the 2017 research cruise. The file named taxon_codes_2017.csv is an associated taxon code list of the observed species.This dataset was transformed from the native format into a table structure using Darwin Core term names as column names.

This reference contains tabular datasets resulting from the eDNA pilot study on National Wildlife Refuges. ZIP file contains all datasets as received from the authors: a folder for each participating refuge containing two Excel workbooks, one for the MiFish marker results and one for the COI marker results. Each workbook has several sheets including one for the raw compiled data, one for each site, and filtered combined data. CSV of filtered data for all participating refuges combined. This dataset was compiled by extracting the filtered datasheet for each refuge from the excel workbook and combining them into a CSV using an r script. CSV of the total OTU, OTU species, unique families, and number of fish, mammal, amphibian, mollusk, and bird species for each participating refuge. This csv was compiled by Rachel Maxey (I&M Data Manager) by extracting the data from the refuge workbooks and combining manually into a CSV. CSV of the full Site data download from Survey 123. Data dictionaries and metadata for site information and eDNA results tables.

File List hb-migration.r (MD5: 1904c1692a02d984890e4575d0eeb4e6) R script that imports the eBird, map, and equal-area icosahedron data, summarizes the population-level migration patterns, runs the statistical analyses, and outputs figures.

    migration-fxns.r (MD5: a2ae2a47c066a253f18cad5b13cddcf6)
     R script that holds the relevant functions for executing the hb-migration.R script.

    BBL-Appendix.r (MD5: 370c701d6afb07851907922dcab51de4)
     R script that imports the Breeding Bird Laboratory data and outputs the figures for the Appendix.

    output-data.zip (MD5: 36e3a92a7d35e84b299d82c8bd746950)
    Folder containing the partially-processed text files (15 .txt files, 3 per species for centroids, migration dates, and migration speed) for the main analyses and figures in the paper. These text files can be used in part II of hb-migration.r and contain output data on the daily population-level centroids, migration dates, and migration speed. Part I of hb-migration.r relies on raw eBird data, which was queried from the eBird server directly. The raw eBird data can be requested through their online portal after making a user account (http://help.ebird.org/customer/portal/articles/1010524-can-i-download-raw-data-from-ebird-). The equal-area icosahedron maps are available at (http://discreteglobalgrids.org/). The BBL data, used in BBL-Appendix.R, can be requested from the USGS Bird Banding Laboratory (http://www.pwrc.usgs.gov/BBL/homepage/datarequest.cfm).

  Description
    The code and data in this supplement allow for the analyses and figures in the paper to be fully replicated using a data set of manipulated communities collected from the literature.
   Requirements: R 3.x, and the following packages: chron, fields, knitr, gamm4, geosphere, ggplot2, ggmap, maps, maptools, mapdata, mgcv, plyr, raster, reshape2, rgdal, Rmisc, SDMTools, sp, spaa, and files containing functions specific to this code (listed above).
    The analyses can then be replicated by changing the working directory at the top of the 
     file hb-migration.R to the location on your computer where you have stored the .R 
    and .csv files and running the code. Note that to fully replicate the analyses, the data will need to be requested from the sources listed above.
    Starting at Part II in hb-migration.R, it should take approximately 30 minutes to run all the code from start to finish. Figures should output as pdfs in your working directory. If you download the raw data and run the analyses starting at Part I, you will need a workstation with large memory to run the analyses in a reasonable amount of time since the raw eBird datafiles are very large.
    Version Control Repository: The full version control repository for this project (including post- publication improvements) is publicly available https://github.com/sarahsupp/hb-migration. If you would like to use the code in this Supplement for your own analyses it is strongly suggested that you use the equivalent code in the repositories as this is the code that is being actively maintained and developed.
    Data use: Partially-processed data is provided in this supplement for the purposes of replication. If you wish to use the raw data for additional research, they should be obtained from the original data providers listed above.

This data provides county level occurrence information for all individuals used in modelling potential exposure and spread of highly pathogenic avian influenza (HPAIv) from the 2021-2022 North American outbreak. The data set contains individual identifiers and taxa information, an indicator of exposure, exposure status (Susceptible, Exposed by HPAIv detection in the county, or Exposed by secondary contact with an exposed bird), and date of first occurrence of each individual bird and that bird's exposure status within each visited county. Herein, county refers to any county, parish, borough, census area, or geographic region identified in the associated geospatial data US_CAN_AI.shp (ESRI shapefile format). Occurrence was determined using a spatial join procedure between GPS relocations of individuals and this geospatial dataset.

[Superseded] This dataset is a single layer from [Superseded] City Plan 2014 – v19.00–2020 collection. Not all layers were updated in this amendment, for more …Show full description[Superseded] This dataset is a single layer from [Superseded] City Plan 2014 – v19.00–2020 collection. Not all layers were updated in this amendment, for more information on past Adopted City Plan amendments. This feature class is shown on the Airport environs overlay map - Bird and bat strike zone and Public safety (map reference: OM-001.4).This feature class includes the following sub-categories:(a) Public safety area sub categories:(i) public safety area sub-category.For information about the overlay and how it is applied, please refer to the Brisbane City Plan 2014 document. Additional information to assist with cross referencing the Airport environs overlay datasets is available in the City Plan 2014 — Airport Environs overlay — referencedataseton Open Data website.The Airport environs overlays contain information derived from data that is created or owned by BAC and licensed to Brisbane City Council. Its use by any person for purposes not associated with planning and development in Brisbane is not authorised.This dataset utilises Brisbane City Council's Open Spatial Data website to provide additional features for viewing and downloading the data.The first resource is in HTML format. The GO TO button will launch our Open Spatial Data website and this will let you preview the data and enable additional download options. The resources labelled GeoJSON, KML and SHP will give you a download of the entire dataset. The ESRI REST resource connects to metadata for the layer while the CSV resource will download attribute data in a table. For more information on the new features and other tips and tricks please read our Blog.

Understanding geographic ranges and species distributions is crucial for effective conservation, especially in the light of climate and land use change. However, the spatial, temporal and intraspecific resolution of digital accessible information on species distributions is often limited. Here, we suggest to make better use of high-resolution tracking data to address existing limitations of occurrence records such as spatial biases (e.g. lack of observations in parts of the geographic range), temporal biases (e.g. lack of observations during a certain period of the year), and insufficient information on intraspecific variability (e.g. lack of population- or individual-level variation). Addressing these gaps can improve our knowledge on geographic ranges, intra-annual changes in species distributions, and population-level differences in habitat and space use. We demonstrate this with tracking data and species distribution models (SDMs) of the Barnacle Goose, a migratory bird species wint..., Tracking data was obtained from the Movebank Data Repository using the following studies:Â

van der Jeugd, H. P., Oosterbeek, K., Ens, B. J., Shamoun-Baranes, J., & Exo, K. (2014). Data from: Forecasting spring from afar? Timing of migration and predictability of phenology along different migration routes of an avian herbivore [Barents Sea data]. [dataset]. https://doi.org/10.5441/001/1.ps244r11Â Heim, W., Piironen, A., Heim, R. J., Piha, M., Seimola, T., Forsman, J. T., & Laaksonen, T. (2022). Data from: Effects of multiple targeted repelling measures on the behaviour of individually tracked birds in an area of increasing human-wildlife conflict [Csv]. Movebank Data Repository. https://doi.org/10.5441/001/1.VD7JB526Â Griffin, L. (2014). Data from: Forecasting spring from afar? Timing of migration and predictability of phenology along different migration routes of an avian herbivore [Svalbard data]. Movebank Data Repository. https://doi.org/10.5441/001/1.5K6B1364Â Garthe, S. (20..., , # Data from: Making better use of tracking data can reveal the spatiotemporal and intraspecific variability of species distributions

https://doi.org/10.5061/dryad.zw3r228fd

Description of the data and file structure

The uploaded data file contains processed tracking data from tracking datasets available for public download via Movebank. Using the package "movepub" (Desmet 2023) to Darwin Core standards, tracking data is sub-sampled to the first record for each hour.

Columns "license", "license holder" and "datasetID" provide information on the original tracking data set, giving the license under which it was published, the contact person for the study, and the DOI (if available respectively). In the absence of a "license", "license holder" or doi, "NA" values are provided, representing "not available".

Columns "institutionCode", "collectionCode" and "datasetName" provide information on the institute responsible for data storage (MPIAB, ...

[Superseded] This dataset is a single layer from [Superseded] City Plan 2014 – v17.00–2019 collection. Not all layers were updated in this amendment, for more information on …Show full description[Superseded] This dataset is a single layer from [Superseded] City Plan 2014 – v17.00–2019 collection. Not all layers were updated in this amendment, for more information on past Adopted City Plan amendments. This feature class is shown on the Airport environs overlay map - Bird and bat strike zone and Public safety (map reference: OM-001.4).This feature class includes the following sub-categories:(a) Bird and bat strike zone sub-categories:(i) distance from airport 0-3km sub-category;(ii) distance from airport 3-8km sub-category;(iii) distance from airport 8-13km sub-category.For information about the overlay and how it is applied, please refer to the Brisbane City Plan 2014 document. Additional information to assist with cross referencing the Airport environs overlay datasets is available in the City Plan 2014 — Airport Environs overlay — reference dataset on Open Data website.The Airport environs overlays contain information derived from data that is created or owned by BAC and licensed to Brisbane City Council. Its use by any person for purposes not associated with planning and development in Brisbane is not authorised.This dataset utilises Brisbane City Council's Open Spatial Data website to provide additional features for viewing and downloading the data.The first resource is in HTML format. The GO TO button will launch our Open Spatial Data website and this will let you preview the data and enable additional download options. The resources labelled GeoJSON, KML and SHP will give you a download of the entire dataset. The ESRI REST resource connects to metadata for the layer while the CSV resource will download attribute data in a table. For more information on the new features and other tips and tricks please read our Blog.

DEEP-VOICE: Real-time Detection of AI-Generated Speech for DeepFake Voice Conversion

This dataset contains examples of real human speech, and DeepFake versions of those speeches by using Retrieval-based Voice Conversion.

Can machine learning be used to detect when speech is AI-generated?

Introduction

There are growing implications surrounding generative AI in the speech domain that enable voice cloning and real-time voice conversion from one individual to another. This technology poses a significant ethical threat and could lead to breaches of privacy and misrepresentation, thus there is an urgent need for real-time detection of AI-generated speech for DeepFake Voice Conversion.

To address the above emerging issues, we are introducing the DEEP-VOICE dataset. DEEP-VOICE is comprised of real human speech from eight well-known figures and their speech converted to one another using Retrieval-based Voice Conversion.

For each speech, the accompaniment ("background noise") was removed before conversion using RVC. The original accompaniment is then added back to the DeepFake speech:

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2039603%2F921dc2241837cd784329955d570f7802%2Fdfcover.png?generation=1692897655324630&alt=media" alt="Overview of the Retrieval-based Voice Conversion process to generate DeepFake speech with Ryan Gosling's speech converted to Margot Robbie. Conversion is run on the extracted vocals before being layered on the original background ambience.">

(Above: Overview of the Retrieval-based Voice Conversion process to generate DeepFake speech with Ryan Gosling's speech converted to Margot Robbie. Conversion is run on the extracted vocals before being layered on the original background ambience.)

Dataset

There are two forms to the dataset that are made available.

First, the raw audio can be found in the "AUDIO" directory. They are arranged within "REAL" and "FAKE" class directories. The audio filenames note which speakers provided the real speech, and which voices they were converted to. For example "Obama-to-Biden" denotes that Barack Obama's speech has been converted to Joe Biden's voice.

Second, the extracted features can be found in the "DATASET-balanced.csv" file. This is the data that was used in the below study. The dataset has each feature extracted from one-second windows of audio and are balanced through random sampling.

**Note: ** All experimental data is found within the "KAGGLE" directory. The "DEMONSTRATION" directory is used for playing cropped and compressed demos in notebooks due to Kaggle's limitations on file size.

A potential use of a successful system could be used for the following:

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2039603%2F7ae536243464f0dbb48f3566765f6b50%2Fdfcover.png?generation=1692897790677119&alt=media" alt="Usage of the real-time system. The end user is notified when the machine learning model has processed the speech audio (e.g. a phone or conference call) and predicted that audio chunks contain AI-generated speech.">

(Above: Usage of the real-time system. The end user is notified when the machine learning model has processed the speech audio (e.g. a phone or conference call) and predicted that audio chunks contain AI-generated speech.)

Papers with Code

The dataset and all studies using it are linked using Papers with Code

The Papers with Code page can be found by clicking here: Papers with Code

Attribution

This dataset was produced from the study "Real-time Detection of AI-Generated Speech for DeepFake Voice Conversion"

Bird, J.J. and Lotfi, A., 2023. Real-time Detection of AI-Generated Speech for DeepFake Voice Conversion. arXiv preprint arXiv:2308.12734.

The preprint can be found on ArXiv by clicking here: Real-time Detection of AI-Generated Speech for DeepFake Voice Conversion

License

This dataset is provided under the MIT License:

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

*THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT H...

From 2010-2018 we investigated the occurrence of avian influenza virus in wild birds in Iceland. A total of 6635 swabs samples were collected from wild birds or fecal material directly associated with wild birds. We screened all samples by a real time - polymerase chain reaction (RT-PCR) test with 381 testing positive. Further testing of all RT-PCR positive samples and all negative samples collected in 2012 by virus isolation yielded 120 positives, with 92 of those testing positive by RT-PCR for avian influenza virus.

This data set contains information regarding the sampling of avian influenza viruses from mallard ducks at locations in Anchorage and Fairbanks, Alaska 2012-2015. Data pertaining to wild birds (mallards) sampled includes band numbers, age and sex, location and timing of sampling. Laboratory specific data is also included and used to identify presence and absence of avian influenza viruses either during active infection or previous exposure (serostatus).

This Zenodo record contains two test datasets (Birds and Littorina) used in the paper:

PhenoLearn: A user-friendly Toolkit for Image Annotation and Deep Learning-Based Phenotyping for Biological Datasets

Authors: Yichen He, Christopher R. Cooney, Steve Maddock, Gavin H. Thomas

PhenoLearn is a graphical and script-based toolkit designed to help biologists annotate and analyse biological images using deep learning. This dataset includes two test cases: one of bird specimen images for semantic segmentation, and another of marine snail (Littorina) images for landmark detection. These datasets are used to demonstrate the PhenoLearn workflow in the accompanying paper.

Dataset Structure

Bird Dataset

• train/ — 120 bird specimen images for annotation and model training.
• pred/ — 100 images for prediction and testing.
• seg_train.csv — Pixel-wise segmentations (CSV format with RLE or polygon masks).
• name_file_pred — Filenames corresponding to prediction images.

Littorina Dataset

• train/ — 120 snail images for training landmark prediction models.
• pred/ — 100 snail images for model testing.
• pts_train.csv — Ground-truth landmark coordinates for training images.
• name_file_pred — Prediction image filenames for evaluation.

How to Use These Datasets

Workflow Instructions (via PhenoLearn)

1. Download the dataset folders.

2. Use PhenoLearn to load seg_train.csv (segmentation) or pts_train.csv (landmark) to view and edit annotations.

3. Train segmentation or landmark prediction models directly via PhenoLearn's training module, or export data for external tools.

4. Use name_file_pred to match predictions with ground-truth for evaluation.

See the full tutorial and usage guide in the https://github.com/EchanHe/PhenoLearn.

A dataset of annotated recordings of a captive colony of rooks, recorded in Strasbourg, France in 2020 and 2021. Each rook was individually identifiable with leg rings. All recordings were taken in the morning a few hours after sunrise, when the birds were most vocally active. The colony was housed outdoors, so other noises are present, including both biotic (most notably various birds, human voices, and other animals) and abiotic (mostly car and train noises).

Audio files (.wav): recorded at 48 kHz, 16-bit using 1 to 3 Song Meter 4 recorders (Wildlife Acoustics). Each recorder had two microphone with different gains to maximise dynamic range. The files were then manually synchronised and merged into multichannel (2 to 6) files.

Label files (.tsv): Labels corresponding to each recording (each pair has the same name), noting the time stamps and individual emitter for each vocalisation. A single observer annotated all the recordings. Only rook vocalisations from the captive colony were annotated, not other bird vocalisations or the various noises in the data. The annotations consist of tables with 5 columns:

Source: the individual producing the vocalisation. Note that only the bird's name is indicated. "Inc" and "Pls" are special cases: the first was for when identity could not be determined, the second when multiple individuals vocalised at once in such a manner that individuals could not be separated

Start: starting time point for the vocalisation, in seconds (determined as the earliest point when the vocalisation was heard on any channel)

End: ending time point for the vocalisation, in seconds (determined as the last point when the vocalisation was head on any channel)

Event: gives information for the bird's activity at the time of the vocalisation, but largely in abbreviated form. One particular case is "sing", which correspond to vocalisations part of a song bout (which are defined as sequences of different vocalisations separated by less than approximately 10 seconds).

Comment: other observations regarding the vocalisation. These are usually not standardised compared to the Event column. One special case is for "Pls": the Comment column then bears information regarding the identity of the individuals involved.

This dataset was used in our article "Acoustic detection and identification of individual rooks in field recordings using multi-task neural networks", to train neural networks to identify individual rooks. The dataset was therefore randomly split into train-validation-test datasets. For reproducibility, we provide the "splitting.csv" which contains the information pertaining to which files go in each dataset, and two scripts to do the split automatically.

To do so: download and unpack the RookID folder somewhere on your computer, then download splitting.csv and either of the scripts to the same location. Both scripts will MOVE, not copy, the files to new folders corresponding to each dataset.

with split_data.R: open the scrip in an RStudio environment, edit the out_path variable to the desired location, and run the script

with split_data.py: run the following command line: python /path/to/split_data.py --out_path path/to/desired/location (note that the script will automatically create the necessary tree structure)

Both scripts can be run without editing the out_path variables, in which case the new folders will be created at the same location

For further information, see our code at https://gitlab.com/kimartin/rook-vocalisation-detection

For any inquiries, please contact Killian Martin (killian.martin@ens-lyon.fr)

This dataset contains the digitized bird atlas data from "Les oiseaux de Sénégambie" (Morel & Morel, 1990). The data consists of the presence data for all species on a degree grid cells in Senegal and Gambia seperatly up to the date of redaction. The book presents itself as an update of "Liste commentée des oiseaux du Sénégal et de la Gambie" (Morel, 1972) and (Morel, 1980).

The code used to generate this dataset is available at https://github.com/Rafnuss/Digitization-of-Oiseaux-de-Senegambie. You can download this dataset as an csv format without all the DwC field at https://github.com/Rafnuss/Digitization-of-Oiseaux-de-Senegambie/tree/main/data

The book can be found on [worldcat.org](https://search.worldcat.org/title/1392421968) or consulted on this [webpage](https://horizon.documentation.ird.fr/exl-doc/pleins_textes/divers20-07/31877.pdf).

EUCA dataset description Associated Paper: EUCA: the End-User-Centered Explainable AI Framework Authors: Weina Jin, Jianyu Fan, Diane Gromala, Philippe Pasquier, Ghassan Hamarneh Introduction: EUCA dataset is for modelling personalized or interactive explainable AI. It contains 309 data points of 32 end-users' preferences on 12 forms of explanation (including feature-, example-, and rule-based explanations). The data were collected from a user study on 32 layperson participants in the Greater Vancouver city area in 2019-2020. In the user study, the participants (P01-P32) were presented with AI-assisted critical tasks on house price prediction, health status prediction, purchasing a self-driving car, and studying for a biological exam [1]. Within each task and for its given explanation goal [2], the participants selected and rank the explanatory forms [3] that they saw the most suitable. 1 EUCA_EndUserXAI_ExplanatoryFormRanking.csv Column description:

Index - Participants' number Case - task-explanation goal combination accept to use AI? trust it? - Participants response to whether they will use AI given the task and explanation goal require explanation? - Participants response to the question whether they request an explanation for the AI 1st, 2nd, 3rd, ... - Explanatory form card selection and ranking cards fulfill requirement? - After the card selection, participants were asked whether the selected card combination fulfill their explainability requirement.

2 EUCA_EndUserXAI_demography.csv It contains the participants demographics, including their age, gender, educational background, and their knowledge and attitudes toward AI. EUCA dataset zip file for download More Context for EUCA Dataset [1] Critical tasks There are four tasks. Task label and their corresponding task titles are: house - Selling your house car - Buying an autonomous driving vehicle health - Personal health decision bird - Learning bird species Please refer to EUCA quantatative data analysis report for the storyboard of the tasks and explanation goals presented in the user study. [2] Explanation goal End-users may have different goals/purposes to check an explanation from AI. The EUCA dataset includes the following 11 explanation goals, with its [label] in the dataset, full name and description

[trust] Calibrate trust: trust is a key to establish human-AI decision-making partnership. Since users can easily distrust or overtrust AI, it is important to calibrate the trust to reflect the capabilities of AI systems.

[safe] Ensure safety: users need to ensure safety of the decision consequences.