This plugin provides a tool for performing sentiment Analysis on textual data.
The plugin comes with a single recipe that allows you to estimate the sentiment polarity (positive/negative) of a text based on it content. For instance, you could use this recipe on a dataset of user reviews or social media data (such as tweets) to know which instances are positive (1) and which are negative (0).
|Author||Dataiku (Hicham EL BOUKKOURI)|
|License||Apache Software License|
How To Use
First, make sure your text data is stored in a dataset, inside a text column, with one row for each document. Using the recipe is very straightforward. Just plug in your dataset, select the column containing the documents you are interested in scoring and run the recipe!
You can tick the “Output confidence scores” box to output the model’s confidence for each prediction.
This plugin uses the text classification library fastText. If you are interested in learning more about how fastText can be used for text classification, you can refer to the following Notebook tutorial.
- P. Bojanowski, E. Grave, A. Joulin, T. Mikolov, Enriching Word Vectors with Subword Information
- A. Joulin, E. Grave, P. Bojanowski, T. Mikolov, Bag of Tricks for Efficient Text Classification
- A. Joulin, E. Grave, P. Bojanowski, M. Douze, H. Jégou, T. Mikolov, FastText.zip: Compressing text classification models
The English model bundled with this plugin comes from fastText, which is BSD-licensed