Next-Word Prediction Project

Introduction

• Objective:
  Create a Shiny app to predict the next word based on an input phrase.

• Deliverables:

  • An interactive Shiny app hosted on shinyapps.io.
    
  • A 5-slide presentation explaining the app and algorithm.

• Applications:

  • Typing assistance tools (e.g., predictive text).
    
  • Enhanced user experience for messaging platforms.

Algorithm Overview

• Approach:
  • Built on a n-gram model that predicts the next word based on prior context.
    
  • The model uses a dataset of common word sequences (phrases) to compute probabilities.
• Preprocessing Steps:
  • Text normalization: lowercasing, tokenization, and removal of special characters.
    
  • Creation of an n-gram dataset to map phrases to probable next words.

Shiny App Features

• Interactive UI:
  • Input a phrase in the text box.
    
  • Press “Predict” to generate the next word.
• Real-Time Predictions:
  • Minimal latency for quick results.
    
  • Handles phrases of varying lengths.
• Deployment:
  • Hosted on shinyapps.io for public access.

Usage Instructions

1. Open the Shiny app using the provided link.
   
2. Enter a phrase in the input box (e.g., “how are”).
   
3. Click the “Predict Next Word” button.
   
4. The predicted word will appear in the output section.

Conclusion

• Key Strengths:
  • Simple, user-friendly interface.
    
  • Real-time next-word predictions.
• Future Enhancements:
  • Incorporate a larger dataset for improved accuracy.
    
  • Upgrade to transformer-based models for contextual understanding.
• Potential Applications:
  • Typing tools, chatbots, and AI-driven messaging systems.

Thank you! Questions?