NRC Lexicon In Python

Overview of NRC Lexicon

The National Research Council of Canada creates an overview of the NRC Lexicon. As discussed earlier, it is one of the most used sentiment lexicons. It provides a vast collection of English words and their associated sentiment values or scores. Also, it becomes essential to understand the structure and the format of the NRC Lexicon and the categories that it classifies.

Explanation of the NRC Lexicon dataset

The NRC Lexcion is a dataset that consists of 14000 English words or phrases. Each of those phrases is labeled with sentiment categories. It helps us understand the sentiment or the emotional tone of the authors writing the text.

It helps us know the emotions and if the sentence is positive or negative, i.e., the sentiment of the phrase. This helps us understand the analysis and produce a more accurate sentiment classification.

Structure and format of NRC Lexicon

NRC Lexicon is stored in the Tab Separated values (TSV) file format. In this file, each row represents a word and the associated sentiment categories. The columns in the TSV file are as follows:

• Word: It contains the 27000 English words for which the sentiment categories are given.
   
• Emotion Categories: Emotion Categories: This category covers the eight basic emotions: Anger, Anticipation, Disgust, Fear, Joy, Sadness, Surprise, and Trust. Each emotion has a corresponding binary sentiment value of 0 and 1.
   
• Sentiment Categories: Apart from the emotions category, we also have a sentiment category that says if the text phrase is negative or positive. The output is again a binary, i.e., 0 and 1, which tells if the sentence is a positive or negative sentiment.

Implementing NRC Lexicon in Python

Introduction to Python libraries for sentiment analysis

Python has various powerful packages on the internet. These can be used to do sentimental analysis. Some of the most popular libraries include NLTK, TextBlob, and VanderSentiment. In this particular sentient of the bog, we will specifically focus on the NRC Lexicon in Python, which is present in the NLTK Library.

Loading and preprocessing the NRC Lexicon dataset

When pursuing a project on sentiment analysis, we need first to load and preprocess the data. As discussed, the NRC Lexicon is in a tab-separated values (TSV) file. We use the Pandas library from Python to load the dataset into the DataFrame.

pip install nrclex

You can also try this code with Online Python Compiler

Run Code

Sentiment analysis using NRC Lexicon in Python

from nrclex import NRCLex

# Create an instance of NRCLex using the input text
text = "This movie is amazing! I absolutely loved it."
sentiment = NRCLex(text)

# Get sentiment scores
positive_score = sentiment.affect_frequencies['positive']
negative_score = sentiment.affect_frequencies['negative']

# Determine the overall sentiment based on scores
if positive_score > negative_score:
    print("Positive sentiment")
elif negative_score > positive_score:
    print("Negative sentiment")
else:
    print("Neutral sentiment")

You can also try this code with Online Python Compiler

Run Code

Output:

Positive sentiment

In this example code, we created an instance of NRCLex module from the Python Library using the input text. This module has all the code required to tokenize the text and then it matches with the NRC Lexicon, thus calculating sentiment scores. We can also understand the sentiment scores with the help of the affect_frequencies  attribute, which will help us to know the frequencies of positive and negative categories.

Emotional analysis using NRC Lexicon in Python

To do the emotional analysis using NRC Lexicon, where it will calculate the frequencies of different emotions present in the text using the lexicon

from nrclex import NRCLex

# Create an instance of NRCLex using the input text
text = "This movie is amazing! I absolutely loved it."
emotions = NRCLex(text).affect_frequencies

# Print the emotions and their frequencies
for emotion, frequency in emotions.items():
    print(emotion, frequency)

You can also try this code with Online Python Compiler

Run Code

Output:

anticipation 0.0
joy 0.6666666666666666
trust 0.0
fear 0.0
surprise 0.0
sadness 0.0
disgust 0.0
anger 0.0

In this example code, we first created an instance of NRCLex using the input text. Then we accessed the emotional frequencies using the affect_frequencies attribute, which is basically a dictionary with emotion names as keys and their related frequencies as values.

The above code snippet prints the emotions and their frequencies present in the input text. In this case, the input text expresses a positive sentiment and is related to the emotion of "joy" with a frequency of approximately 0.67 which so the highest of all. Other emotions have frequencies of 0 since they are not expressed in the given text.

Limitations and challenges of NRC Lexicon

There are some limitations to this excellent technique which is used for sentiment analysis:

1. Words Coverage: The NRC Lexicon does not include all the words, and also it does not take into consideration the surrounding words which provide the context of the whole sentence. It needs to be more precise when determining the context of the sentence.
    
2. Accurate Predictions: Though it provides the usual positive and negative labels, but it doesn’t tell us about the intensity, sarcasm, or cultural variations. This limits the accuracy of the output that it gives.

Frequently Asked Questions

What is sentiment analysis?

Sentiment analysis is extracting opinions from the text. It determines whether the sentiment expressed is positive, negative, or neutral.

What is the NRC Lexicon?

The NRC Lexicon is a dataset associating words with sentiments. It helps identify positive and negative sentiments in text for sentiment analysis tasks.

How does NRC Lexicon handle negations?

The NRC Lexicon doesn't handle negations directly, but by detecting negation words like "not," sentiment polarity can be reversed for words following the negation to capture negated sentiments.

Conclusion

In conclusion, we have seen in this blog the NRC Lexicon in Python and how it is so important in sentiment analysis. We took an overview of sentimental analysis, NRC Lexicon in Python, and how it is implemented in the coding part to understand the emotions or sentiments of the text. Lastly, we also saw the limitations and challenges faced in NRC Lexicon.

If you would like to explore Python and its packages more, refer to the blogs below for more information

• Python Packages
   
• Modules and Packages in Python
   
• Popular Python Libraries
   
•   reverse a string in python
   

For more information, refer to our Guided Path on Coding Ninjas Studio to upskill yourself in Python, Data Structures and Algorithms, Competitive Programming, System Design, and many more! Head over to our practice platform, Coding Ninjas Studio, to practice top problems, attempt mock tests, read interview experiences and interview bundles, follow guided paths for placement preparations, and much more!