Hi R fraternity,
I am trying to compare two text docs in R but not able to get the results. I am using Minhashing and locality sensitive hashing.
Reprex of the code is-
library(textreuse)
library(shiny)
library(RecordLinkage)
library(tidyverse)
library(dplyr)
library(shinydashboard)
library(tesseract)
library(magick)
library(stringdist)
ui<- fluidPage(
fluidRow(
column(4,
fileInput("image1", "Upload Image 1")),
column(4,
fileInput("image2", "Upload Image 2")),
column(8,
actionButton("submit", "Submit"),
actionButton("compare", "Analyze"))
),
fluidRow(column(12,
box(width = 12,
uiOutput("text"),
title = "Text Output 1"
)
)),
fluidRow(column(12,
box(width = 12,
uiOutput("text1"),
title = "Text Output 2"
)
)),
fluidRow(column(12,
box(width = 12,
verbatimTextOutput("analysis"),
title = "Analysis"
)
)))
server<- function(input, output, session)
{
text1<- reactive({
t1<- c(image_ocr(image_read(input$image1, path = input$image1$datapath)))
t<- as.character(t1)
return(t1)
})
text2<- reactive({
t2<-c(image_ocr(image_read(input$image2, path = input$image2$datapath)))
t2<-as.character(t2)
return(t2)
})
observeEvent(input$submit, {
req(input$image1)
output$text<- renderUI({
text1()
})
})
observeEvent(input$submit, {
req(input$image2)
output$text1<- renderUI({
text2()
})
})
a<- reactive({
t<-c(text1(),text2())
minhash <- minhash_generator(200, seed = 235)
ats <- TextReuseCorpus(text = t,
tokenizer = tokenize_ngrams, n = 7,
minhash_func = minhash)
buckets <- lsh(ats, bands = 50, progress = FALSE)
candidates <- lsh_candidates(buckets)
scores <- lsh_compare(candidates, ats, jaccard_similarity, progress = FALSE)
return(scores)
})
observeEvent(input$compare,{
output$analysis<- renderPrint({
req(a())
})
})
}
shinyApp(ui,server)
o\p I am getting-