About us

We strongly believe that data is essential. Our mission is to popularize crowd science within the ML and DS community and build up crowdsourcing experts (crowd solutions architects).
Why master crowdsourcing?
  • Data is the key to success. If you have bad quality data you won’t get usable results.
  • Crowdsourcing provides fast and effective data labeling for complex machine learning tasks.
  • For research, a crowdsourcing platform enables fast and cost-efficient scaling with better reliability in results.
  • Methodology is essential for controlling quality. The quality of data labeling depends on how data is collected and how the task is defined — not on the skills of individual performers.
  • According to research, data scientists spend the majority of their time preparing data. Crowdsourcing helps to optimize this process.
What we offer
Online course
An introductory course on the methodology and practical applications of crowdsourcing for data collection and labeling.
Start learning
Crowd Science Seminar
A venue where researchers and practitioners from across the globe can discuss their past and future work, exchange ideas, and establish new collaborations.
Knowledge base
Find success with our helpful guides and step-by-step instructions.
Python library
We have an open-sourced library with a client that covers all API functionalities.
Online course
An introductory course on the methodology and practical applications of crowdsourcing for data collection and labeling.
Crowd Science Seminar
A venue where researchers and practitioners from across the globe can discuss their past and future work, exchange ideas, and establish new collaborations.
Crowdscience.ai
A space to bring together the research and industry communities to develop the practices and policies of reliable data collection — a backbone of fair, trustworthy, and efficient AI.
Python library
Crowd-kit is a Python library that implements most standard crowdsourcing algorithms. It consists of various methods for aggregating performers’ answers, metrics to estimate answer and performer quality, and quality control techniques. The library has an easy-to-use scikit-learn-like interface and operates with Pandas data frames.
Knowledge base
Find success with our helpful guides and step-by-step instructions.
Application for corporate training
We offer corporate training to help you solve challenges and develop an internal team of crowd science architects (CSA).
YouTube channel
Watch short and clear video tutorials, free webinars, and training courses. Learn how to work with the crowd, manage data labeling, and get quality results.
Blog
Explore updates to the platform, case studies, and articles on cutting-edge technologies.

Partnerships

Grant program
Universities
Coming soon
Referral program
Job Opportunities

Our educational events

Online course at CUSO
Our course at the Conférence universitaire de Suisse occidentale introduces crowdsourcing as a practical methodology and helps participants master the essential steps and techniques to ensure top-quality data.

7th - 25th June 2021

Online Y-DATA Community Course
Together with Y-DATA in Tel Aviv, we created a 6-week community course taught by field experts. Students are exposed to both theory and practice, and apply their new knowledge in 5 full-cycle hands-on projects.

21st April - 26th May 2021

Tutorial at the Academic Fringe Festival
We introduce data labeling via public crowdsourcing marketplaces and present the key techniques for efficiently collecting labeled data. In the practice session, participants launch their own labeling project.

April 8th, 2021

Explore, exchange and apply new ideas
Cookie files
Yandex uses cookies to personalize its services. By continuing to use this site, you agree to this cookie usage. You can learn more about cookies and how your data is processed in the Privacy Policy.
Mon Jun 07 2021 15:05:40 GMT+0300 (Moscow Standard Time)