colorized.text.labeling.py

import nltk
from nltk.sentiment import SentimentIntensityAnalyzer
from nltk import pos_tag, word_tokenize
from colour import Color

nltk.download('vader_lexicon')
nltk.download('averaged_perceptron_tagger')
nltk.download('punkt')

def generate_color(sentiment_score, pos_tag):
    """
    Generate a color based on sentiment score and part of speech.
    Uses a gradient for sentiment and modifies the shade based on the part of speech.
    """
    base_color = Color(hue=sentiment_score * 0.5 + 0.5, saturation=1, luminance=0.5)

    # Modify the color based on part of speech
    if pos_tag.startswith('NN'):
        base_color.luminance *= 0.9  # Darken for nouns
    elif pos_tag.startswith('VB'):
        base_color.saturation *= 0.7  # Desaturate for verbs
    # ... other parts of speech adjustments

    return base_color.hex_l

def colorize_text(text):
    sia = SentimentIntensityAnalyzer()
    words = word_tokenize(text)
    tagged_words = pos_tag(words)

    colored_text = ""
    for word, tag in tagged_words:
        sentiment_score = sia.polarity_scores(word)['compound']
        color = generate_color(sentiment_score, tag)
        colored_word = f'<span style="color:{color}">{word}</span>'
        colored_text += colored_word + ' '

    return colored_text

# Example usage
sample_text = "The quick brown fox jumps over the lazy dog."
colorized = colorize_text(sample_text)
print(colorized)