Animal Social Network Repository
This network repository is composed of seven-hundred and seventy networks from sixty-nine different species publicly available. This repository is novel in that it incorporates different animal networks that vary in sociality levels, including social, solitary and fission fusion. Networks vary in interaction criteria, edge definition, and weighted and unweighted, and in many more network properties as well. However, this repository synthesizes all of these networks in an organized, consistent manner in order to make sure that the networks are user-friendly. It is our hope that the networks in this repository further advance the field of wildlife network theory, yet the networks in this repository are not limited to questions of sociality. Due to its extensive and detailed nature, the networks found in this repository can be used to answer multiple scientific inquiries in varying fields.
To learn more about the project, please visit our website: https://bansallab.github.io/asnr/ or read the work by Sah et al cited below.
To cite the repository, please use:
Pratha Sah, José David Mendez, Shweta Bansal. Animal Social Network Repository, Version 1. https://github.com/bansallab/asnr (2019)
Melissa Collier, Sania Ali, Shweta Bansal. Animal Social Network Repository, Version 2. https://github.com/bansallab/asnr (2021)
To cite the project and data, please use:
Pratha Sah, José David Mendez, Shweta Bansal. A multi-species respository of social networks. Scientific Data, 6:44 (2019)
The list of available datasets are formatted in graphml files and are available here:
GraphML Index <https://github.com/jdm286/Animal-Social-Structure-Network-Repository"_
Species: Scientific name of the animals recorded in the study.
Taxonomic class: Taxonomic class of the species. Could be Actinopterygii, Aves, Insecta, Mammalia or Reptilia.
Population type: Type of animal population recorded. Could be captive, semi-ranging or free-ranging.
Georgraphical location: Location where the population(s) were recorded.
Data collection technique: Method of data collection including focal sampling, survey scan, mark recapture, RFID, radio tags, video recording, etc.
Interaction type: Type of association/interaction recorded between animal pairs to construct the social network.
Definition of interaction: Criteria of occurence of an association/interaction as defined in the study.
Edge weight type: Criteria of weighting network edges. Could be based on frequency, duration, intensity of interactions or metrics such as simple ratio index, half weight index etc.
Total duration of data collection: Total time span over which the social network data was collected.
Time resolution of data colllection (within a day): Resolution of data collection within a day.
Time span of data collection (within a day): Total duration of data collection within a day.
Attributes Available: The node attributes provided in the study. If available, they are included in the graphml files
Note: Additional notes on the dataset
One of the main goals of this repository is to further enhance the study of animal social networks. If you happen to come across any interesting data on animal networks that has been published and is not yet included in this repository, please contact us. We will do our best to make it consistent with our format and add it to the repository. Requests should be filed to shweta@sbansal.com
All datasets in this repository were made publicly available in other data repositories. You will find a link to the original source of the network, additionally to the study in which the data was collected. A great amount of effort was made to determine the license under which the datasets were distributed, and it is our understanding that the datasets are free to redistribute. However, if you own the rights to any dataset that is included in this repository and you object their inclusion, please email shweta@sbansal.com. The data in question will promptly be removed as well as all traces from git revision history.