[R] Mechanical Turk vs alternatives for Data Labeling

Hello Redditors, I'm part of an academic lab and we're looking to annotate 10,000s of images using human labelers. There are a few options available, such as:

* Amazon Mechanical Turk
* Amazon SageMaker
* Appen
* Clarifai
* many more

It seems like Amazon Mechanical Turk (MTurk) is the fastest to get started with, since you can just start creating tasks using their web interface. All the other services require a whole quoting and proposal process with a team …

academic amazon amazon sagemaker appen clarifai data data labeling human images lab labeling machinelearning part process sagemaker services team web