Final_Assigniment

# Load necessary libraries
library(shiny)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(stringr)

# Load or create a pre-trained n-gram model
# Replace this placeholder with your trained model or data
# For demonstration purposes, we create a dummy n-gram dataset
ngram_data <- data.frame(
  phrase = c("how are", "how is", "how do", "what is", "what are"),
  prediction = c("you", "it", "we", "this", "they"),
  stringsAsFactors = FALSE
)

# Define UI
ui <- fluidPage(
  titlePanel("Next-Word Prediction App"),
  sidebarLayout(
    sidebarPanel(
      textInput("phrase", "Enter a phrase:", ""),
      actionButton("predict", "Predict Next Word")
    ),
    mainPanel(
      h4("Prediction"),
      verbatimTextOutput("outputPrediction")
    )
  )
)

# Define server
server <- function(input, output) {
  # Reactive expression for prediction
  predict_next_word <- reactive({
    req(input$phrase)  # Ensure input is provided
    phrase <- tolower(str_trim(input$phrase))
    match <- ngram_data %>%
      filter(str_detect(phrase, paste0("^", phrase))) %>%
      slice(1)  # Take the first match
    if (nrow(match) > 0) {
      match$prediction
    } else {
      "No prediction available"
    }
  })

  # Output prediction
  output$outputPrediction <- renderText({
    input$predict  # Trigger on button click
    isolate(predict_next_word())
  })
}

# Run the application
shinyApp(ui = ui, server = server)

Shiny applications not supported in static R Markdown documents