📌 SMS Spam–Ham Classification Dataset

📖 Overview

The SMS Spam–Ham Classification Dataset contains 5,574 SMS messages, each labeled as either:

• Ham → Legitimate / normal message
• Spam → Unwanted / promotional message

This dataset is widely used for text classification, spam filtering, and NLP model evaluation.

---

📊 Dataset Summary

Label	Count	Percentage
Ham	4,827	86.6%
Spam	747	13.4%
Total	5,574	100%

---

Format

• Each line contains:

  1. Label → ham or spam
  2. Message text → Raw SMS content

Example:

ham   What you doing? How are you?  
ham   Ok lar... Joking wif u oni...  
spam  FreeMsg: CALL to No: 86888 & claim your reward...  
spam  URGENT! Your Mobile No 07808726822 was awarded a bonus...  

🛠️ Usage

This dataset can be used for:

• Building spam detection models.

• Experimenting with NLP techniques (tokenization, TF-IDF, embeddings).

• Benchmarking ML algorithms such as:

  • Naive Bayes
  • SVM
  • Random Forest
  • Deep Learning models

---

🌍 Applications

• SMS Spam Filtering on mobile devices.
• Email Spam Detection (conceptually similar).
• Text Mining & NLP Research.