VADER_SENTIMENT

Overview

The VADER_SENTIMENT function performs sentiment analysis on text using the VADER (Valence Aware Dictionary and sEntiment Reasoner) lexicon, which is specifically attuned to sentiments expressed in social media. VADER  is a rule-based model that combines a sentiment lexicon with five heuristics to account for punctuation, capitalization, degree modifiers, conjunctions, and negation. For more details, see the official VADER documentation .

This implementation uses the NLTK  package, which provides a Python interface to the VADER algorithm. For more information, see the NLTK documentation .

The final sentiment is calculated as a normalized, weighted composite score called the compound score, which ranges from -1 (most negative) to +1 (most positive). This score is computed as follows:

where $x$ is the sum of valence scores and $\alpha$ is a normalization constant (default $\alpha = 15$).

This example function is provided as-is without any representation of accuracy.

Usage

To use the function in Excel:

=VADER_SENTIMENT(text)

• text (string, required): The text to analyze.

The function returns a single value (float): the compound sentiment score between -1 (most negative) and +1 (most positive), or 0.0 for invalid or empty input.

Examples

Example 1: Positive Product Review

Input:

Excel formula:

=VADER_SENTIMENT("I absolutely love this product! It exceeded all my expectations and I would highly recommend it to anyone.")

Expected output:

Example 2: Negative Customer Feedback

Input:

Excel formula:

=VADER_SENTIMENT("The customer service was terrible. I waited for hours and my issue was never resolved properly.")

Expected output:

Example 3: Neutral Business Statement

Input:

Excel formula:

=VADER_SENTIMENT("Our company will be conducting the annual inventory count on June 15th. All departments will participate as scheduled.")

Expected output:

Example 4: Mixed Sentiment

Input:

Excel formula:

=VADER_SENTIMENT("While the product quality is excellent, the shipping was delayed which was disappointing.")

Expected output:

Python Code

import nltk
from nltk.sentiment import SentimentIntensityAnalyzer
nltk.download('vader_lexicon')
 
def vader_sentiment(text):
    """
    Analyze sentiment of text using VADER.
 
    Args:
        text: The text to analyze (string).
 
    Returns:
        The compound sentiment score (float) between -1 (most negative) and +1 (most positive), or 0.0 for invalid or empty input.
 
    This example function is provided as-is without any representation of accuracy.
    """
    if not isinstance(text, str) or not text.strip():
        return 0.0
    sia = SentimentIntensityAnalyzer()
    sentiment = sia.polarity_scores(text)
    return round(sentiment['compound'], 4)