• A Shiny app that predicts the next word in a phrase
  
• Built using an n-gram language model in R
  
• Designed to be fast, simple, and user-friendly

• Predictive text helps users type faster
  
• Used in messaging, email, and search engines
  
• Improves user experience and efficiency
  
• Goal: build a simple and practical prediction tool

• Training data from blogs, news, and Twitter
  
• Text cleaned by removing punctuation, links, and symbols

Model approach:
- Uses n-grams: bigrams, trigrams, fourgrams

• User enters a phrase
  
• App cleans the input text
  
• Model predicts the next word using backoff:
  • 4-word → 3-word → 2-word → fallback

Example:
- Input: “I love”
- Output: “you”

• Fast and responsive
  
• Simple interface
  
• Clear prediction logic
  
• Scalable with more data

Future improvements:
- Better smoothing
- Larger datasets
- Multiple word predictions