Network Dataset

An anonymized version of the Pinterest social networks and activities used in our WWW14 and ICWSM13 papers is being made available to the research community. If you are interested in using this data, please send us an email according to the Request Data section and indicate which of following parts you need in the email.

1. Pinterest network: A snowball sampled social graph of Pinterest, crawled in Apr 2013.

2. Facebook network: The Facebook social graphs of users who appear in the Pinterest activites dataset below.

3. Fb-copied network: The subset of Pinterest network that only contains links common to both Pinterest and Facebook.

4. User information: The basic statistics (such as the number of pins, likes, followees and followers) of Pinterest users.

5. Pinterest activities: Repin and like activity in Pinterest during 03-21 Jan, 2013.

6. Pinterest pins: Basic information (e.g. the source of image) and statistics (e.g. number of repins and likes) of 3.36 million images published during 03-21 Jan, 2013.

You can find the format of the dataset from here.

Image Dataset

An anonymized version of the Pinterest image dataset used in our WWW15 paper is being made available to the research community. If you are interested in using this data, please send us an email according to the contact us section and indicate which of following parts you need in the email.

1. Image features: The object, deep learning and image quality features of images in our dataset. (Refer the table 1 of our WWW15 paper for details.)

2. User features: The profiles and category preference of users in our datase.

3. Repin activity: The repin activities in our dataset, including the repinned category and pinboards.

4. Images URLs: The URL of images in our dataset collected in 03-21 Jan, 2013.

You can find the format of the dataset from here.

Request Data

If you are interested in using this data, please fill the form to Request Data to get the link where you can download the data.

Terms and Conditions

1. You will use the data solely for the purpose of non-profit research or non-profit education.

2. You will respect the privacy of end users and organizations that may be identified in the data. You will not attempt to reverse engineer, decrypt, de-anonymize, derive or otherwise re-identify anonymized information.

3. You will not distribute the data beyond your immediate research group.

4. If you create a publication using our datasets, please cite our papers as follows.

   • If you use the 1-4 parts of the Network Dataset, please cite our WWW14 paper.

   • If you use the 5-6 parts of the Network Dataset, please cite our ICWSM15 paper.

   • If you use the Image Dataset, please cite our WWW15 paper.

@inproceedings{www2015zhong,
  author = {Changtao Zhong, Dmytro Karamshuk and Nishanth Sastry},
  title = {{Predicting Pinterest: Automating a Distributed Human Computation}},
  booktitle = {{24th International World Wide Web Conference (WWW)}},
  month = {May},
  year = {2015},
  address = {Florence, Italy}
}

@inproceedings{www2014zhong,
  author = {Changtao Zhong and Mostafa Salehi and Sunil Shah and Marius Cobzarenco and Nishanth Sastry and Meeyoung Cha},
  title = {{Social Bootstrapping: How Pinterest and Last.fm Social Communities Benefit by Borrowing Links from Facebook}},
  booktitle = {{23rd International World Wide Web Conference (WWW)}},
  month = {April},
  year = {2014},
  address = {Seoul, Korea}
}

@inproceedings{icwsm2013zhong,
  title = {Sharing the Loves: Understanding the How and Why of Online Content Curation},
  booktitle = {7th International {AAAI} Conference on Weblogs and Social Media ({ICWSM13)}},
  author = {Zhong, Changtao and Shah, Sunil and Sundaravadivelan, Karthik and Sastry, Nishanth},
  year = {2013},
  month = {July},
  address = {Boston, US}
}