The Flow
Start by looking at the flow. It begins with a folder containing the reviews to be classified in raw form, downloaded from this page. A Python recipe turns these reviews into a Dataiku dataset.
EXPLORE !The goal of the project is to classify text from IMDB reviews by positive or negative sentiment.
Start by looking at the flow. It begins with a folder containing the reviews to be classified in raw form, downloaded from this page. A Python recipe turns these reviews into a Dataiku dataset.
EXPLORE !The plugin includes a recipe for classifying text with a trained model. It takes as inputs a column of text to classify. The output dataset contains the predictions and the confidence in those predictions.
EXPLORE !