Upload 2 files

app.R

@@ -1,51 +1,235 @@
+# Import necessary libraries
 library(shiny)
-library(bslib)
-library(dplyr)
+library(caret)
+library(readr)
+library(catboost)
 library(ggplot2)
 
-df <- readr::read_csv("penguins.csv")
-# Find subset of columns that are suitable for scatter plot
-df_num <- df |> select(where(is.numeric), -Year)
+# Load the pre-trained model
+model <- readRDS("goat_behavior_model_caret.rds")
 
-ui <- page_fillable(theme = bs_theme(bootswatch = "minty"),
-  layout_sidebar(fillable = TRUE,
-    sidebar(
-      varSelectInput("xvar", "X variable", df_num, selected = "Bill Length (mm)"),
-      varSelectInput("yvar", "Y variable", df_num, selected = "Bill Depth (mm)"),
-      checkboxGroupInput("species", "Filter by species",
-        choices = unique(df$Species), selected = unique(df$Species)
-      ),
-      hr(), # Add a horizontal rule
-      checkboxInput("by_species", "Show species", TRUE),
-      checkboxInput("show_margins", "Show marginal plots", TRUE),
-      checkboxInput("smooth", "Add smoother"),
+# Define UI for application
+ui <- fluidPage(
+  
+  # App title ----
+  titlePanel("Detecting Goat Behaviors"),
+  
+  # Sidebar layout with input and output definitions ----
+  sidebarLayout(
+    
+    # Sidebar panel for inputs ----
+    sidebarPanel(
+      
+      # Input: Select a file ----
+      fileInput("file1", "Choose TSV File",
+                accept = c(
+                  "text/tsv",
+                  "text/tab-separated-values,text/plain",
+                  ".tsv")
+      )
+      
     ),
-    plotOutput("scatter")
+    
+    # Main panel for displaying outputs ----
+    mainPanel(
+      
+      # Output: Tabset with data, confusion matrix, and download button
+      tabsetPanel(
+        id = "dataset",
+        tabPanel("About", 
+                 HTML("
+                 <h5> The following model was part of the the research article: 
+                 <h4>Developing an Interpretable Machine Learning Model for the Detection of Mimosa Grazing in Goats</h4>
+                 
+                 <em>In the last years, several machine learning approaches for detecting animal behaviors have been proposed. 
+                 However, despite their successful application,  their complexity and lack of explainability have difficulty in their 
+                 application to real-world scenarios. The article presents a machine-learning model for differentiating between grazing mimosa and other activities  
+                 (resting, walking, and grazing ) in goats using sensor data.  Boruta, an algorithm for selecting the most relevant features, and SHAP, 
+                 a technique for interpreting the decision of a machine learning model are two fundamental components of the methodology used for creating the model. 
+                 The resulting model, a gradient boost algorithm with 15 selected features proved to be extremely accurate in detecting Grazing activities.  
+                 The study demonstrates the fundamental role of model explainability in identifying model weaknesses and errors, thereby creating a path for future 
+                 improvements. In addition, the simplicity of the resulting model not only reduces computational complexity and processing time but also enhances 
+                 interpretability and facilitates the deployment of real-life scenarios.</em>
+                 <p>
+                 <p>This application allows users to test the pre-trained machine learning model that predicts goat behavior based on input sensor data. 
+                       The input data should be a tab-separated value (.tsv) file containing specific sensor data related to the goat's activity.
+                       <p>The application then generates predictions, provides a confusion matrix result, and offers the option to download the predictions.
+                       <p>The key features expected in the dataset are: 
+                      <table>
+  <thead>
+    <tr>
+      <th>No</th>
+      <th>Feature</th>
+      <th>Definition</th>
+    </tr>
+  </thead>
+  <tbody>
+    <tr>
+      <td>1</td>
+      <td>Steps</td>
+      <td>Number of steps</td>
+    </tr>
+    <tr>
+      <td>2</td>
+      <td>HeadDown</td>
+      <td>% time with head down</td>
+    </tr>
+    <tr>
+      <td>3</td>
+      <td>Standing</td>
+      <td>% time Standing</td>
+    </tr>
+    <tr>
+      <td>4</td>
+      <td>Active</td>
+      <td>% time Active</td>
+    </tr>
+    <tr>
+      <td>5</td>
+      <td>MeanXY</td>
+      <td>Arithmetic mean between X and Y positions</td>
+    </tr>
+    <tr>
+      <td>6</td>
+      <td>Distance</td>
+      <td>Distance in meters</td>
+    </tr>
+    <tr>
+      <td>7</td>
+      <td>prev_steps1</td>
+      <td>Number of steps one step backward</td>
+    </tr>
+    <tr>
+      <td>8</td>
+      <td>X_Act</td>
+      <td>X position actuator</td>
+    </tr>
+    <tr>
+      <td>9</td>
+      <td>prev_Active1</td>
+      <td>% time Active one step backward</td>
+    </tr>
+    <tr>
+      <td>10</td>
+      <td>prev_Standing1</td>
+      <td>% time Standing one step backward</td>
+    </tr>
+    <tr>
+      <td>11</td>
+      <td>DFA123</td>
+      <td>Accumulative Euclidean distance from actual position to three positions forward</td>
+    </tr>
+    <tr>
+      <td>12</td>
+      <td>prev_headdown1</td>
+      <td>% time with head down one step backward</td>
+    </tr>
+    <tr>
+      <td>13</td>
+      <td>Lying</td>
+      <td>% time Lying</td>
+    </tr>
+    <tr>
+      <td>14</td>
+      <td>Y_Act</td>
+      <td>Y position actuator</td>
+    </tr>
+    <tr>
+      <td>15</td>
+      <td>DBA123</td>
+      <td>Accumulative Euclidean distance from actual position to three positions backward</td>
+    </tr>
+  </tbody>
+</table> 
+                      
+                      <p><p> <h5> Experiments, source code and more <a href=https://github.com/harpomaxx/goat-behavior/> here<a/> </h5>")
+        ),
+        
+        tabPanel("Results",
+                 tableOutput("contents"),
+                 verbatimTextOutput("confusionMatText"),
+                 plotOutput("confusionMatPlot"),
+                 downloadButton("downloadData", "Download Predictions"))
+      )
+    )
   )
 )
 
-server <- function(input, output, session) {
-  subsetted <- reactive({
-    req(input$species)
-    df |> filter(Species %in% input$species)
+# Define server logic 
+server <- function(input, output) {
+  
+  # For the predictions dataset
+  predictions <- reactive({
+    
+    if (is.null(input$file1))
+      return(NULL)
+    
+    inFile <- input$file1
+    dataset <- readr::read_delim(inFile$datapath,delim='\t')
+    predict(model, dataset)
+    
   })
   
-  output$scatter <- renderPlot({
-    p <- ggplot(subsetted(), aes(!!input$xvar, !!input$yvar)) + list(
-      theme(legend.position = "bottom"),
-      if (input$by_species) aes(color=Species),
-      geom_point(),
-      if (input$smooth) geom_smooth()
-    )
-
-    if (input$show_margins) {
-      margin_type <- if (input$by_species) "density" else "histogram"
-      p <- p |> ggExtra::ggMarginal(type = margin_type, margins = "both",
-        size = 8, groupColour = input$by_species, groupFill = input$by_species)
+  # For the table
+  output$contents <- renderTable({
+    
+    # input$file1 will be NULL initially. After the user selects
+    # a file, it will be a data frame with 'name', 'size', 'type', and 'datapath' variables.
+    inFile <- input$file1
+    
+    if (is.null(inFile))
+      return(NULL)
+    
+    # Read the file from the input$file1 path and return it
+    dataset <- readr::read_delim(inFile$datapath,delim='\t')
+    head(dataset, n = 5)
+  })
+  
+  # Download function for predictions
+  output$downloadData <- downloadHandler(
+    filename = function() {
+      paste("predictions-", Sys.Date(), ".csv", sep="")
+    },
+    content = function(file) {
+      write.csv(data.frame(Index = 1:length(predictions()), Prediction = predictions()), file, row.names = FALSE)
     }
+  )
+  
+  # Confusion Matrix
+  output$confusionMatText <- renderPrint({
+    
+    if (is.null(input$file1))
+      return(NULL)
     
-    p
-  }, res = 100)
+    inFile <- input$file1
+    dataset <- readr::read_delim(inFile$datapath,delim='\t',progress = FALSE)
+    predictions <- predict(model, dataset)
+    cm<-caret::confusionMatrix(reference=as.factor(dataset$Activity),predictions,mode="everything")
+    cm$overall
+  })
+  
+  output$confusionMatPlot <- renderPlot({
+    
+    if (is.null(input$file1))
+      return(NULL)
+    
+    inFile <- input$file1
+    dataset <- readr::read_delim(inFile$datapath,delim='\t')
+    predictions <- predict(model, dataset)
+    cm<-caret::confusionMatrix(reference=as.factor(dataset$Activity),predictions,mode="everything")
+    
+    # Extract table data from confusion matrix
+    confusionMatrixTable <- as.table(cm$table)
+    
+    # Plot the confusion matrix
+    ggplot(as.data.frame(confusionMatrixTable), aes(x=Reference, y=Prediction)) +
+      geom_tile(aes(fill = log(Freq)), colour = "white") +
+      geom_text(aes(label = sprintf("%1.0f", Freq)), vjust = 1) +
+      scale_fill_gradient(low = "white", high = "steelblue") +
+      theme_minimal() +
+      theme(axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1))
+  })
 }
 
-shinyApp(ui, server)
+# Create a Shiny app object
+shinyApp(ui = ui, server = server)