interactive-task-map.Rmd

---
output: html_document
runtime: shiny
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)

library(ggplot2)
library(e1071)
library(plotly)
library(dplyr)
library(tidyverse)
```

```{r load-data, echo=FALSE, message=FALSE, warning=FALSE}
task_map <- read_csv('./task_map.csv')
```
```{r interactive_map, echo = F}
task_map_pca <- prcomp(task_map[, -1], scale = TRUE)
task_map_pca_result <- as.data.frame(task_map_pca$x[, 1:2])
task_map_pca_result$TaskName <- task_map$task

# Define the UI
ui <- fluidPage(
  titlePanel("Interactive Task Map"),
  sidebarLayout(
    sidebarPanel(
      numericInput("k_value", "Enter the value of k:", min = 1, value = 2),
      actionButton("cluster_btn", "Apply Clustering"),
      actionButton("reset_btn", "Reset Clustering")
    ),
    mainPanel(
      plotlyOutput("map_plot")
    )
  )
)

# Define the server
server <- function(input, output) {
  # Initialize clustering data frame
  clustering_data <- reactiveValues(data = NULL)

  # Apply k-means clustering
  applyClustering <- function(k) {
    if (k > 1) {
      cluster_result <- kmeans(task_map[, -1], centers = k)
      clustering_data$data <- task_map_pca_result
      clustering_data$data$cluster <- as.factor(cluster_result$cluster)
    }
  }

  # Default clustering
  default_clustering <- reactive({
    applyClustering(2)  # Set the default value of k
    clustering_data$data
  })

  # Apply k-means clustering when the button is clicked
  observeEvent(input$cluster_btn, {
    k <- input$k_value
    applyClustering(k)
  })

  # Reset clustering
  observeEvent(input$reset_btn, {
    clustering_data$data <- task_map_pca_result
    clustering_data$data$cluster <- NULL
  })

  # Plotly map plot
  output$map_plot <- renderPlotly({
    plot_data <- clustering_data$data

    ggplotly(
      ggplot(plot_data, aes(x = PC1, y = PC2, color = cluster, text = TaskName)) +
        geom_point(size = 3) +
        theme_minimal()
    )
  })

  # Update default clustering when task_map_pca_result changes
  observe({
    if (is.null(clustering_data$data)) {
      clustering_data$data <- default_clustering()
    }
  })
}

# Run the Shiny app
shinyApp(ui, server)
```