Aggregation: A key step in data labeling

During the webinar, we will discuss various types of aggregation algorithms and learn how to use them in the Toloka interface and in Python.

Image
Image

Overview

Methods for aggregating and drawing meaningful inferences from the noisy labels obtained from the crowd are the cornerstone of crowdsourcing. During the webinar, we will discuss various types of aggregation algorithms and learn how to use them in the Toloka interface and in Python.

In this webinar, you will learn:
  • Why we need aggregation
  • How different models work
  • Which method to use

Speaker

Image
Nikita Pavlichenko
TolokaNikita is a Research Scientist at Toloka. He focuses on quality control and aggregation of crowdsourcing responses. He is an M.S. student at the Moscow Institute of Physics and Technology and previously worked on protein structure prediction in collaboration with the Inria NANO-D team.

Recording

Don't miss out

Be the first to hear about our workshops, 
tutorials, and webinars.
Fractal