Sentiment Analysis

Sentiment Analysis Approaches

Approach	How It Works	Pros	Cons
Lexicon-based (VADER)	Pre-built word sentiment scores	No training needed, fast	Misses context and sarcasm
ML Classifier	Trained Naive Bayes / SVM	Learns domain patterns	Needs labeled training data
Deep Learning (BERT)	Fine-tuned transformer model	State-of-the-art accuracy	Compute intensive

from nltk.sentiment.vader import SentimentIntensityAnalyzer
from transformers import pipeline
import nltk

nltk.download('vader_lexicon', quiet=True)
sia = SentimentIntensityAnalyzer()

reviews = [
    "This product is absolutely amazing!",
    "Terrible quality. Broke after 2 days.",
    "It's okay I guess. Nothing special.",
]

# VADER (fast, rule-based)
for review in reviews:
    scores = sia.polarity_scores(review)
    label = ("POSITIVE" if scores['compound'] >= 0.05
             else "NEGATIVE" if scores['compound'] <= -0.05 else "NEUTRAL")
    print(f"[{label}] {scores['compound']:+.3f} | {review}")

# Transformer (accurate, context-aware)
classifier = pipeline("sentiment-analysis")
results = classifier(reviews)
for review, r in zip(reviews, results):
    print(f"{r['label']} ({r['score']:.3f}) → {review[:50]}")