https://github.com/draperunner/fjlc

A Python port of the Fredriksen–Jahren Lexicon Classifier
https://github.com/draperunner/fjlc

idi ntnu sentiment sentiment-analysis sentiment-classification twitter twitter-sentiment-analysis

Last synced: 19 days ago
JSON representation

A Python port of the Fredriksen–Jahren Lexicon Classifier

• Host: GitHub
• URL: https://github.com/draperunner/fjlc
• Owner: draperunner
• License: mit
• Created: 2017-04-06T16:43:39.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2017-04-17T14:31:15.000Z (almost 8 years ago)
• Last Synced: 2024-12-05T23:34:04.228Z (about 2 months ago)
• Topics: idi, ntnu, sentiment, sentiment-analysis, sentiment-classification, twitter, twitter-sentiment-analysis
• Language: Python
• Size: 38.9 MB
• Stars: 3
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE.txt

Awesome Lists containing this project

README

        
# Lexicon Classifier

Source code: https://github.com/draperunner/fjlc

This package is a Python port of the Lexicon Creator and Classifier of Valerij Fredriksen and Brage Ekroll Jahren (2016).

It is compatible with Python version >= 3.

The original Java code is available here: https://github.com/freva/Masteroppgave

If using this package in your publications, please cite

> Valerij Fredriksen and Brage Ekroll Jahren. Twitter Sentiment Analysis: Exploring Automatic Creation of Sentiment Lexica. Master's thesis, 2016.

## Installation

```bash

pip install fjlc

```

## Lexicon Classifier

The `LexiconClassifier` uses the best performing lexicon of Fredriksen and Jahren. You can specify your own lexicon, see Options below.

### Usage

```python

from fjlc import LexiconClassifier

lc = LexiconClassifier()

```

You can classify a single tweet or a list of tweets:

```

lc.classify("I am happy!")  # 'POSITIVE'

lc.classify(["I am happy!", "I hate rain"])  # ['POSITIVE', 'NEGATIVE']

```

You can get the sentiment value of a single tweet or multiple tweets

```

lc.calculate_sentiment("I am happy!")  # 5.599244615570646

lc.calculate_sentiment(["I am happy!", "I hate rain"])  # [5.599244615570646, -2.767224666516315]

```

### Options

The `LexiconClassifier` takes three options:

* `lexicon`: Path to sentiment lexicon file

* `options`: Path to options file

* `dictionary`: Path to canonical dictionary

## Lexicon Creator

### Usage

```python

from fjlc import LexiconCreator

lc = LexiconCreator()

```

Incomplete, untested.