🛠 Mini Project: Spam Classifier with Machine Learning

🛠 Mini Project: Spam Classifier with Machine Learning

In this mini project, we'll build a simple Spam Classifier using Machine Learning techniques. We’ll use TF-IDF Vectorization for feature extraction and a Naive Bayes Classifier for classification.

🔹 Step 1: Import Required Libraries

We'll need scikit-learn library, which provides all the tools we need.

🔹 Step 2: Prepare the Data

We create a small dataset with labels "spam" and "ham" (not spam).

🔹 Step 3: Build and Train the Model

We will transform the text messages into numerical vectors using TF-IDF and train a Naive Bayes classifier.

# Step 1: Import Libraries
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.naive_bayes import MultinomialNB
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

# Step 2: Create Dataset
texts = [
    "Congratulations! You've won a free iPhone!",
    "Your loan is approved, call now!",
    "Let's meet for lunch tomorrow.",
    "Important update about your account.",
    "Win cash prizes easily!",
    "Are you attending the meeting today?"
]
labels = ["spam", "spam", "ham", "ham", "spam", "ham"]

# Step 3: Preprocess and Split Data
X_train, X_test, y_train, y_test = train_test_split(texts, labels, test_size=0.3, random_state=42)

# Step 4: Feature Extraction
vectorizer = TfidfVectorizer()
X_train_vectors = vectorizer.fit_transform(X_train)
X_test_vectors = vectorizer.transform(X_test)

# Step 5: Train Model
model = MultinomialNB()
model.fit(X_train_vectors, y_train)

# Step 6: Predictions
predictions = model.predict(X_test_vectors)

# Step 7: Accuracy
print("Accuracy:", accuracy_score(y_test, predictions))

# Step 8: Testing on New Message
test_message = ["Win an amazing prize now!"]
test_vector = vectorizer.transform(test_message)
result = model.predict(test_vector)
print("Prediction:", result[0])

🔹 Sample Output:

Accuracy: 1.0
Prediction: spam

---