• Brief Introduction to the Problem:
  • Predicting the next word in a sequence of text.
• Importance:
  • Enhancing text prediction accuracy has wide-ranging applications in natural language processing.
• Purpose:
  • Introduce our innovative solution.

• Overview of the Algorithm Used:
  • We utilized n-gram models for next word prediction.
• Explanation of N-gram Models:
  • N-gram models analyze sequences of N words to predict the next word.
  • Modern methods and libraries (tokenizers, SnowballC) was used to create n-grams

# Read and preprocess text data
text <- readLines("pg73352.txt")
clean_text <- tolower(text)
clean_text <- gsub("[[:punct:]]", "", clean_text)
clean_text <- clean_text[clean_text != ""]

# Tokenize the text into unigrams, bigrams, and trigrams
trigrams <- unlist(tokenize_ngrams(clean_text, n = 3))

# Function to build n-gram model
build_ngram_model <- function(ngram_data) {
  ngram_freq <- table(ngram_data)
  return(ngram_freq)
}

• Datasets:
  • As dataset was used book https://www.gutenberg.org/cache/epub/73352/pg73352.txt
• Overview of the Shiny App:
  • Our Shiny app allows users to input a phrase and receive a prediction for the next word.
• Instructions:
  • Simply enter your phrase in the text box and click ‘Predict’.
• Demonstration:
  • Screenshots or GIF demonstrating how the app functions.

• User Experience:
  • Users experience seamless prediction with our app, improving text composition efficiency.
  • You can test in by clicking the link below: https://7346it-nikita-sivonin.shinyapps.io/PredictNextWordApp/
• Conclusion:
  • Our app offers a user-friendly interface and accurate predictions, paving the way for enhanced text analysis.