Goal of the Project

• Objective: Build a Shiny app that predicts the next word based on a given phrase.
• Key Features:
  • Takes a phrase as input (multiple words).
  • Predicts the next word using a bigram model.
  • Displays the predicted next word as the output.

What is a Bigram Model?

• A bigram is a sequence of two adjacent words in a text.
• The model predicts the next word based on the previous word.
• We use the probability of the next word given the current word in the text.

Model Training

• Tokenized the text into words.
• Created bigrams (pairs of consecutive words).
• Calculated the frequency and probability of the next word based on previous words.

Workflow

1. User Input: The user enters a phrase (multiple words).
2. Preprocessing: The app extracts the last word from the input.
3. Prediction: Using the bigram model, the app predicts the most probable next word.
4. Output: The predicted next word is displayed to the user.

Why is this Algorithm Effective?

• Simple and Efficient: The bigram model is a simple and efficient way to predict the next word based on the most recent word.
• Real-Time Prediction: The Shiny app provides real-time prediction, offering an interactive user experience.
• Scalable: This approach can be extended to more complex models like trigrams or even neural networks.

Future Improvements

• Multiple Predictions: Expand the app to show the top N predicted words.
• Better Model: Consider using a trigram or neural network model for better predictions.
• User Customization: Allow users to provide custom training data for personalized predictions.

Conclusion

• The Next Word Prediction App offers an interactive and user-friendly interface for predicting the next word in a given phrase.
• It leverages a simple bigram model to provide useful, real-time predictions for users.