Practical Implementation: Building a Simple Sentiment Classifier

Advanced

😊 Sentiment Analysis with Scikit-learn

Let's implement a basic sentiment analysis model using Python and scikit-learn.
The goal: classify movie reviews as positive or negative.

🛠️ Sample Code: Movie Review Classifier

from sklearn.datasets import load_files
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.model_selection import train_test_split
from sklearn.naive_bayes import MultinomialNB
from sklearn.metrics import accuracy_score

# Load dataset
reviews = load_files('reviews')
X = reviews.data
y = reviews.target

# Preprocess text
vectorizer = TfidfVectorizer(stop_words='english', max_df=0.7)
X_vectorized = vectorizer.fit_transform(X)

# Split data
X_train, X_test, y_train, y_test = train_test_split(X_vectorized, y, test_size=0.2, random_state=42)

# Train classifier
clf = MultinomialNB()
clf.fit(X_train, y_train)

# Evaluate
predictions = clf.predict(X_test)
print(f'Accuracy: {accuracy_score(y_test, predictions):.2f}')

🔍 Key Components

  • 🗂️ Dataset Loading: Loads labeled movie reviews
  • 🧹 TF-IDF Vectorization: Converts raw text to numerical features
  • 🔀 Train/Test Split: Evaluates generalization performance
  • 🤖 Multinomial Naive Bayes: A simple, effective text classifier
  • 📊 Accuracy Score: Measures classification success

🚀 Why It Matters

This pipeline demonstrates how to quickly develop and evaluate an NLP classifier — a foundational workflow for many sentiment analysis tasks.

🧩 Diagram: Sentiment Analysis Pipeline

Movie Reviews
      |
      v
 [TF-IDF Vectorizer]
      |
      v
[Naive Bayes Classifier]
      |
      v
  Sentiment Prediction
 (Positive / Negative)