Creating a Dataset Sucks. Here's What I've Learned to Make it a Little Bit Easier

Written by calmdownkarm | Published 2020/08/16
Tech Story Tags: deep-learning | data-science | annotation-automation | dataset | automation | machine-learning | creating-a-dataset-sucks | creating-a-dataset

TLDR Creating datasets is hard, even for relatively simple tasks like document classification/sentiment analysis. Even a relatively high agreement score can allow for 10% of your data to be wrong. Hiring a set of annotators is hard too, but once you figure out the hiring, here's some things that I've learned managing annotators. Getting a minimal set of data annotated, and trying to build models on it will help target the specific kinds of annotated data that you need, or even bad classifiers can help filter through a lot of data.via the TL;DR App
no story
Published by HackerNoon on 2020/08/16
ABOUTHow hackers start their afternoons. We publish tech stories and build publishing software.

READEntirely free library read by hundreds of millions to learn anything about any technology.

WRITEHackerNoon elevates quality technology writing far and wide across the interwebs.

PARTNERHackerNoon works directly with tech companies to place ads by content relevancy