Next Word Prediction Application

Merana

2024-09-17

Slide 1: Title and Introduction

Title: Next Word Prediction Application
Introduction: This project involves predicting the next word based on the input text using n-gram models.
Objectives: Develop a Shiny app that predicts the next word and visualize predictions with probabilities.

Slide 2: Description of the Algorithm

N-Gram Model: Utilizes unigram, bigram, and trigram models to predict the next word.
Algorithm:
- Trigrams: Matches the last two words and predicts the next word.
- Bigrams: Matches the last word and predicts the next word.
- Unigrams: Returns the most frequent words if no other matches are found.

Slide 3: Application Overview

Shiny Application: Interactive tool for predicting the next word.
Features:
- Text Input: Users can enter text to get predictions.
- Reset Button: Clears the input field.
- Predicted Words: Displays the most probable next word(s).
Screenshot:

Slide 4: How the Application Works

Functionality:
- Users enter a phrase, and the app predicts the next word based on n-gram models.
- The app calculates probabilities and displays them in a bar chart.
Example Usage:
- Input: “happy new”
- Output: “year, york” (predicted next words)

Slide 5: Conclusion and Evaluation

Functionality Review:
- The app effectively predicts the next word and provides probabilities.
- User-friendly interface with clear instructions and results.
Future Improvements:
- Enhance prediction accuracy.
- Expand to handle more complex queries.
Evaluation:
- The app demonstrates a solid understanding of predictive modeling and Shiny application development.
Try it!
- https://merana-tech.shinyapps.io/shiny_app2/