# Sentiment Analysis

**Supported Labeling Types**: `Row labeling`

Sentiment analysis will provide a simple classification between “Positive” or “Negative”.

The model used for sentiment analysis is sourced from [DistilBERT-base-uncased-finetuned-SST-2](https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english). This model is a fine-tuned checkpoint of DistilBERT-base-uncased, specifically trained on the SST-2 dataset.

<figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-8d47614ef03144385abdc3d1b2e47f6bea7bc1f1%2FExtension%20-%20ML-assisted%20Labeling%20-%20Row%20labeling%20-%20Sentiment%20analysis%20-%20highlight.png?alt=media" alt="Image of ML Assisted with Sentiment Analysis"><figcaption><p>ML Assisted with Sentiment Analysis</p></figcaption></figure>

### Model Details

* The model is using a distilled version developed by Hugging Face based on the Text Classification task pipeline.
* Trained on Stanford Sentiment Treebank ([sst2](https://huggingface.co/datasets/stanfordnlp/sst2)) corpora which contains 67,349 movie review excerpts with human-annotated sentiment labels.
* The model reaches an accuracy of **91.3%** on the development set.
* The model hosted locally within the Datasaur Intelligence container.

### Usage

* This model is primarily used for sentiment classification and also be used for topic classification.
* The raw model supports masked language modeling and next sentence prediction, though it is primarily intended for fine-tuning on downstream tasks.
* To explore additional fine-tuned versions for different tasks, check out the [Hugging Face model hub](https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english).