A New Data Source...

  • 130% Funded!

    CrowdSignals.io funded with for your support

  • Check out our Blog

    For Updates and Information on CrowdSignals.io

  • Crowdfund Your Data

    Help Us Fund Your Next Big Dataset

  • Big Mobile

    The Community's Largest Labeled Mobile Dataset

  • Ethical Data Collection

    Best Practices and Fair Compensation

  • Multi-Device Data

    from Smartphones and Smartwatches


Join 1000s of researchers, data scientists, and businesses in a diversity of fields. Contribute your support and expert feedback by joining the community!


Join the Community for updates and info on participation!


Email Format:   



We're social!

  About CrowdSignals.io  

Building the Community's Largest Labeled Mobile and Sensor Dataset

Ethical Mobile and Sensor Data

CrowdSignals.io will create the largest set of rich, longitudinal mobile and sensor data recorded from smartphones and smartwatches available to the community. The dataset will include geo-location, sensor, system and network logs, user interactions, social connections, and communications as well as user-provided ground truth labels and survey feedback. The data will be collected from a demographically diverse pool of Android users across the United States.

We will not profit from this dataset - all funds raised through crowdfunding will contribute directly to the data collection effort, this includes compensation for staff as well as for volunteers who consent to share their anonymized data. We will use best practices in recruiting, on-boarding, and managing volunteers. For volunteers, this means informed consent, the ability to opt-out at any time, and knowledge of who is using the data and for what purposes. For campaign sponsors, this means a legal agreement to commit to ethical use of the data as well as some transparency regarding how they will use of the data.

Rewards for Sponsors and Volunteers

CrowdSignals.io will generate many terabytes of mobile and sensor data recorded from smartphones and smartwatches over the duration of the data collection period. All sponsors will have access to the core, anonymized data a few weeks after the data is collected. Higher-level sponsors may additionally specify their own demographic surveys or ground truth labels to be used by volunteers recruited to the CrowdSignals.io data collection campaign. The dataset will be made available to non-sponsors for research purposes 12-18 months after the data is collected, at this time it will be free of charge but subject to the same constraints on ethical use.

For volunteers, there will be several educational and financial incentives. First, they will learn about the risks and benefits of sharing mobile data during informed consent. They will receive financial compensation along with access to insights and analytics about their own data after it is collected. Volunteers will also receive regular updates on how their data is enabling advances in a variety of fields from geography, to computer science, mobile health, social science, and many other areas.

How it Works

The Goal

Enable 1000's of Scientists and Developers Across a Variety of Fields

Goal

Our goal is to enable students, researchers, data scientists, and product groups across a variety of fields with the data they need to solve important societal problems - and to collect this data using best practices that inform, engage, and compensate data collection volunteers while respecting privacy.

CrowdSignals.io will use crowdfunding to finance the collection of a massive, shared dataset at a per sponsor cost that is orders of magnitude less than an in-house data collection. We aim to build a large enough group of sponsors to make the data accessible even to students who need data for a thesis or class project (e.g., $1-$2 per data collection volunteer, so a data from 100 volunteers could cost $100).

Status

We've collected a large amount of very useful feedback from academia and industry on priorities for: (1) data collection parameters (e.g., data recorded, collection protocols, participant demographics, device categories), (2) ground truth event labels to collect from participants (e.g., activities, events, situations) and (3) anonymization techniques and data license terms that maximize the utility of the data while offering an adequate level of protection for data collection participants.

In March 2016 we're planning to launch a crowdfunding campaign for Phase 1 of CrowdSignals.io in which we will collect data from 30+ volunteers for 30 days.

Building the Large Hadron Collider
for Internet of Things Data Science

  Sponsors  

Institutions that Sponsor CrowdSignals.io

Goergen Institute for Data Science
University of Rochester

The leading academic sponsor of CrowdSignals.io is the Goergen Institute for Data Science at the University of Rochester. The Goergen Institute supports interdisciplinary research in data science across the University of Rochester’s College of Arts and Sciences, the Hajim College of Engineering and Applied Sciences, the School of Medicine and Dentistry and the University of Rochester Medical Center, and the Eastman School of Music. The Institute offers undergraduate and Masters programs in Data Science. In 2017, the Institute and the Department of Computer Science will move into Wegmans Hall, a state of the art facility now under construction. For more information, see www.rochester.edu/data-science
→ Learn More

Lumme Inc.

Take the last step towards becoming an ex-smoker with Lumme! We help you lead a healthier life by offering the help just when you need it.


→ Learn More

Smart Data Innovation Lab
KIT/TECO

Researchers and industry will be able to analyze CrowdSignals.io user tracking data with a 12TB in-memory big-data platform @SDIL_DE

→ Learn More

CrowdSignals Community

Domain Experts who Lead or Support CrowdSignals.io

Thomas Ploetz
Backer, CrowdSignals.io
Reader
Open Lab, Newcastle University

Unsupervised learning over wearable and ubiquitous sensor data for behaviour analysis


→ Learn More

Rijurekha Sen
Backer, CrowdSignals.io
Post-Doctoral Researcher
Max Planck Institute for Software

Researching mobile privacy for apps that use encrypted data.


→ Learn More

Simon Kamronn
Backer, CrowdSignals.io
Ph.D. Student
Technical University of Denmark

Researching deep learning models of human behavior to support behavior change.


→ Learn More

Abhinav Parate
Backer, CrowdSignals.io
Head of R&D, Lumme Inc.

Developing a personalized quit program for smokers by combining wearable sensors, data analytics, and behavioral psychology


→ Learn More

Adam Joinson
Backer, CrowdSignals.io
Professor of Information Systems
University of Bath

Studying issues of public-private space and activity, mood and use, and patterns of behaviour


→ Learn More

Christian Poellabauer
Backer, CrowdSignals.io
Associate Professor
University of Notre Dame

Studying systems and methods that combine various types and sources of information to develop new or improved context-aware apps and services.
→ Learn More

R Venkatesha Prasad
Backer, CrowdSignals.io
Assistant Professor, TU Delft

"Data is everywhere! But useful and insightful data is nowhere! Look at CrowdSignals.io"


→ Learn More

Mario Parreño-Centeno
Backer, CrowdSignals.io
Ph.D. Student, Newcastle University
EPSRC CDT in Cloud Computing for Big Data

"A dataset which was required. Excellent project!"


→ Learn More

Flora Salim
Backer, CrowdSignals.io
Senior Lecturer, RMIT University

Leveraging heterogeneous sensor data to analyse and predict fine-grained behaviours in human mobility.
→ Learn More

Michael Noll-Hussong
Backer, CrowdSignals.io
MD, University of Ulm

Using personal device data to study mental disorders in combination with social neuroscience.

→ Learn More

Jonathan Rubin
Backer, CrowdSignals.io
Research Scientist, PARC

Deep learning for Pervasive Health and Affective Computing

→ Learn More

Takeshi Okadome
Backer, CrowdSignals.io
Professor, Kwansei Gakuin University

Improving recognition of Activities of Daily Life with more diverse sensor data

→ Learn More

Vinayak Naik
Backer, CrowdSignals.io
Associate Professsor at IIIT-Delhi

Crowdsourcing collective wisdom through mobile and personal devices

→ Learn More

Deborah Estrin
Adviser, CrowdSignals.io
Professor, Computer Science, Cornell Tech
Healthcare Policy and Research, WCMC

"CrowdSignals will provide critical infrastructure for small-data application research that empowers end-users to make use of their own data."






→ Learn More

Henry Tirri
Adviser, CrowdSignals.io
Executive in Residence
Aalto University (Finland)

"Modern technology development is driven by data. CrowdSignals will accelerate innovation in many disciplines for the good of humanity. Data collection campaigns like this are expensive, time consuming and very hard for any single research group to organize - CrowdSignals will provide a competitive edge at an excellent value for researchers both in academia and industry."

→ Learn More

Evan Welbourne
Organizer, CrowdSignals.io
Founder and CEO, AlgoSnap

"We organized CrowdSignals.io to address the critical scarcity of shared mobile data sets. By combining insights, expertise, and small contributions from the community we're creating terabytes of high-quality data that will drive research in a diversity of fields. Data from our devices will eventually help us to understand ourselves as individuals and as a society."

→ Learn More

Jason Hong
Adviser, CrowdSignals.io
Associate Professor, Carnegie Mellon University

"If successful, CrowdSignals.io will produce data sets that accelerate research on HCI, privacy, the Internet of Things, and many other topics"





→ Learn More

Andrew Campbell
Adviser, CrowdSignals.io
Professor, Dartmouth College

"We can infer potentially vital information about ourselves with the data captured by our personal devices - from behavioral trends to our overall health and well-being. Datasets like CrowdSignals are fundamental to advancing research on how we can use our devices to improve our quality of life."



→ Learn More

Anind Dey
Charles M. Geschke Director
Human-Computer Interaction Institute
Carnegie Mellon University

"CrowdSignals is an amazing opportunity to advance research on countless topics, particularly for understanding people through ubiquitous and mobile computing! At a cost that's dramatically lower than in-house data collection campaigns, this will be accessible to everyone including professionals, researchers and students."

→ Learn More

Lama Nachman
Senior Principal Engineer
Director of Anticipatory Computing
Intel Labs

"I am really excited about CrowdSignals because it solves one of the major bottlenecks for algorithmic innovation in context awareness: the lack of sufficient labeled sensor data. This could become the community’s gold standard for benchmarking algorithms and creating new ones"

→ Learn More

Henning Schulzrinne
Levi Professor
Columbia University


"The CrowdSignals project will enable collection of high-fidelity network data at low-cost and unprecedented scale. This could be a big step forward for mobile network research! We'll be able to use the data in numerous studies, for example we could use the network logs to develop models that predict network usage behavior."
→ Learn More

Sam Madden
Professor, MIT CSAIL
Director, bigdata@CSAIL
Co-Founder, Cambridge Mobile Telematics

"This is a super cool idea! CrowdSignals represents a new source of Big Mobile Data which is likely to enable research in a variety of areas."




→ Learn More

Hari Balakrishnan
Fujitsu Chair Professor, MIT CSAIL
Co-Founder and CTO
Cambridge Mobile Telematics

"CrowdSignals is an excellent idea! The data will provide a great way to evaluate new mobile systems and applications."




→ Learn More

Tanzeem Choudhury
Associate Professor
Cornell


"I was really excited to hear about this project. [...] By unifying the interests of the researchers as well as the stakeholders we can potentially reach out to more people and really get the scale of data we need to make mobile sensing really successful and effective across a wide range of people."

→ Watch the Interview

Glen Duncan
Associate Professor
Epidemiology
University of Washington

"This is a potentially great opportunity to collect a wide swath of data in different environmental contexts - I personally could see a lot of use for this data in the kinds of research I normally do with the built environment and the social environment"


→ Watch the Interview

John Shen
Professor, ECE Department
Carnegie Mellon University


"CrowdSignals will enable researchers from diverse disciplines to conduct deep research into human behavior modeling and prediction, and to carry out large-scale in-situ experiments. With the capabilities provided by CrowdSignals, new crowd-sensing and inferencing systems can be developed to provide valuable services to people, private enterprises, and civic organizations."

→ Learn More

Seng Loke
Reader and Associate Professor
La Trobe University (Australia)

"I like the idea of CrowdSignals - such labelled mobile data is not easy to come by, especially lots of it, which is important in many research problems. There is much to learn from the crowd to build new crowd-powered mobile computing systems."




→ Learn More

David Racz
Principal Scientist
Microsoft

"This is a great, unique opportunity to build a data collection platform that serves the needs of thousands of researchers worldwide in a variety of fields of study and at the same time, help provide incentives for people to contribute their data. Crowdsignals will allow the collection of datasets that will have enormous value for all kinds of partners and stakeholders."


→ Watch the Interview

Jesus Favela
Professor
CICESE (Mexico)

"This initiative to collect this large collection of data from mobile users could be really a threshold in research in ubiquitous computing, and particularly, in activity recognition, and behavior recognition, and in finding the ways in which these behaviors or these activities will impact our mood, impact our health, etc."


→ Watch the Interview

Romit Roy Choudhury
Associate Professor
ECE and CS
University of Illinois at Urbana-Champaign

"Collecting sensor data from the wild is often not the difficult problem - the difficulty often emerges from associating ground-truth labels to this data. CrowdSignals is systematically approaching the task of labeling, which can make the data set incredibly useful to a wide variety of applications. Moreover, the collective time savings across academia and industry would be enormous."
→ Learn More

Vidya Setlur
Research Scientist
Tableau Software


"This project will be extremely valuable to a lot of people whether it's small scale - single users, or a company. [...] It would be extremely valuable if the entire data can be collected from a community and then shared, given back to the community - and coming up with those clever ways of empowering the user with his information."

→ Watch the Interview

Andy Hickl, CEO
Alan Liu, Research Scientist
ARO Mobile (Now at Vulcan)


"I can tell you that CrowdSignals is important because it saves my company money - because you've got high-quality data capture, you've got high-fidelity data distribution. But the reason CrowdSignals is most important to me is that it gives the power of all this data to the community."







→ Watch the Interview

Henry Kautz
Robin & Tim Wentworth Director
Goergen Institute for Data Science
University of Rochester

"The Crowdsignals data sets will be invaluable for driving progress on modeling complex human behavior from sensor data. Most research to date has used different data sets, which are expensive to create and are each collected under different protocols. Crowdsignals will lower the bar to entry for new researchers and support true apple-to-apple comparisons of algorithms. Researchers in pervasive and ubiquitous computing, artificial intelligence, and human-computer interaction will find it to be a boon."

→ Learn More

Lin Zhong
Associate Professor
ECE and CS Departments
Rice University

"CrowdSignals will provide a vital source of data on the everyday performance of mobile computing and communication systems for a diversity of users across the United States. Unlike previous efforts pursued by academics like our LiveLab project, CrowdSignals data collection model could scale up to many more users in a cost-effective manner. When the privacy concerns are adequately addressed through anonymization and other techniques, the resulting dataset could be fundamental to our work on future computing and communication systems."
→ Learn More

Dan Ashbrook
Assistant Professor
Information Sciences and Computer Science
Rochester Institute of Technology

"One of the best ways to improve human interaction with mobile devices is to understand what those humans are doing with their devices. CrowdSignals' ability to collect interaction, context, and questionnaire data will help us to build new kinds of interfaces to improve people's everyday lives."





→ Learn More

Aaron Quigley
Chair of Human Computer Interaction
University of St Andrews

"The ethical use and scaleable collection of data is an increasingly challenging problem faced in both academic and industrial research. Conflicting needs and interests are often masked or ignored as data is used, abused and repurposed without respect for the people who own it or created it. This mobile sensing approach in CrowdSignals.io is perfectly placed to address this challenge. Respect for the data matched with respect for the people, while leveraging the power of the crowd to create a global knowledge platform is the key to success here."

→ Learn More

Deepak Jagdish
Research Assistant
MIT Media Lab


"I'm a huge fan of the project because it's going to offer data in many layers - usually it's only one layer, just sensor data or just user interaction data, but here's a chance for us to get data from a lot of users for all of these layers. You could learn very many interesting things from it."






→ Watch the Interview

  Free Sample Datasets  

Smartphone + Smartwatch Data Collected using the CrowdSignals Platform

Download free sample datasets along with code in Java and Python for exploring it.
Many hours of continuous data from 20+ sensors on a smartwatch and smartphone.
Labels for 15 human activities provided by participants.



Download Datasets
CrowdSignals solves the chicken and egg problem for mobile and sensor data

The Data

User-labeled location, sensor, interaction, social, system, and network data collected simultaneously from smartphones and smartwatches

Geo-Location and Radios

Geo-Location and Radios

Location and radios including Bluetooth, GSM, and WLAN. This data is fundamental for work with location, place, mobility patterns, and mHealth.

Read the Details
Social Communication Data

Social and Communications

Calendar, contacts, calls, SMS, and other social data. Fundamental metadata for computational social science and studies of mobile-social behaviors.

Read the Details
System and Networking Data

System and Networking

Logs of system and network use such as battery, current connection, and network transmissions. Required for studies of energy and mobile networking.

Read the Details



User Interaction Data

User Interaction

Logs on app usage, web browsing, media consumption, phone configuration, and other user interaction. Required for studies of smartphone usage and mobile user interaction.

Read the Details
Sensors

Motion and Field Sensors

Sensors including acclerometer, gyroscope, magnetic field, and others. This data is often used in activty recognition and mHealth as well as environmental and situational context.

Read the Details
Survey Data

User Survey Feedback

Rich demographic surveys, experience sampling surveys, and lightweight lockscreen surveys. Adds rich, frequent ground truth for training labels and verification of hypotheses.

Read the Details

FAQ

Frequently Asked Questions

  We maintain an FAQ section where you can find more information about CrowdSignals.io data collection campaign procedures, legal agreements, sponsors and volunteers among other topics.

About AlgoSnap

  AlgoSnap was founded in 2015 to provide services that accelerate algorithm design, evaluation, and deployment in the Internet of Things. AlgoSnap is a C corporation and it is the legal entity to which all contracts, data privacy, and other legal agreements for CrowdSignals.io will refer. AlgoSnap receives advising on CrowdSignals.io from a global panel of experts and works with a major Silicon Valley law firm with world-class specialists in data privacy law. All funds raised by the CrowdSignals.io campaign will contribute directly to the CrowdSignals.io data collection effort, this includes compensation for AlgoSnap staff as well as for volunteers who consent to share their anonymized data. For more on AlgoSnap, see our website.

Press Kit

Please find our high-res infographics here

Please contact us for details!

Copyright © 2015 AlgoSnap Inc. / All rights reserved.