Emotion classification using NRC Lexicon in Python

Many a time, for real-world projects, emotion recognition is often just the start of the project. That time writing a whole code on that will not only increase time but also efficiency is hindered. 

NRCLexicon is an MIT-approved pypi project by Mark M. Bailey which predicts the sentiments and emotion of a given text. The package contains approximately 27,000 words and is based on the National Research Council Canada (NRC) affect lexicon and the NLTK library’s WordNet synonym sets.

Installation:

To install this module type the below command in the terminal. 

pip install NRCLex

Even after the installation of this module, MissingCorpusError may occur while running programs. So it is advised to also install textblob.download_corpora by using the below command on the command prompt. 

python -m textblob.download_corpora

Approach:

• Import the module

Python3

# Import required modules
from nrclex import NRCLex

• Assign input text

Python3

# Assigning list of words
text = ['hate', 'lovely', 'person', 'worst']

• Create NRCLex object for each input text.

Python3

for i in range(len(text)):

    # creating objects
    emotion = NRCLex(text[i])

• Apply methods to classify emotions.

Sr.	Method	Description
1	emotion.words	Return words list.
2	emotion.sentences	Return sentences list.
3	emotion.affect_list	Return affect list.
4	emotion.affect_dict	Return affect dictionary.
5	emotion.raw_emotion_scores	Return raw emotional counts.
6	emotion.top_emotions	Return highest emotions.
7	emotion.affect_frequencies	Return affect frequencies.

• Emotional affects measured include the following:

1. fear
2. anger
3. anticipation
4. trust
5. surprise
6. positive
7. negative
8. sadness
9. disgust
10. joy

Below is the Implementation.

Example 1:

Based on the above approach, the below example classifies various emotions using top_emotions.

Python3

# Import module
from nrclex import NRCLex

# Assign list of strings
text = ['hate', 'lovely', 'person', 'worst']

# Iterate through list
for i in range(len(text)):

    # Create object
    emotion = NRCLex(text[i])

    # Classify emotion
    print('\n\n', text[i], ': ', emotion.top_emotions)

Output:

Example 2:

Here a single emotion love is classified using all the methods of NCRLex module.

Python3

# Import module
from nrclex import NRCLex

# Assign emotion
text = 'love'

# Create object
emotion = NRCLex(text)

# Using methods to classigy emotion
print('\n', emotion.words)
print('\n', emotion.sentences)
print('\n', emotion.affect_list)
print('\n', emotion.affect_dict)
print('\n', emotion.raw_emotion_scores)
print('\n', emotion.top_emotions)
print('\n', emotion.affect_frequencies)

Output: