Skip to main content
Dataset View

Where to start when building MonkeyLearn Workflows

Written by Francisco Decuzzi
Updated over a week ago

This is the main tab, where can quickly explore the dataset, slice and dice it (using the faceting on the left) and search for particular samples.

Each row is a training sample or register.

The text column is a special one that contain the text that will be processed, the rest can be any other metadata (dates, ratings, etc) that come with the original data as inputs.

Columns can also be enrichments that are generated by ML models (eg: sentiment, keywords, topics, etc), see the model view.

There's a special column length, which contains the length of the text in characters. This is a very handy column to help explore the data sorting by the text length.

The main goals of this View are to:

  • Explore the data, understand it and visualize it.

  • Faceting and segmenting data by different values from metadata and/or from model predictions.

  • Understand which samples where labeled, which were not and to understand how to improve precision and coverage.

Did this answer your question?