title: "Iris Species Visualization" runtime: shiny output: flexdashboard::flexdashboard: orientation: rows social: menu sourcecode: embed theme: simplex

```{r setup , include=FALSE} library(shiny) library(plotly) library(leaflet) library(tidyverse) library(sf) library(highcharter) library(kableExtra) df <- datasets::iris

lookup = structure(c("setosa","versicolor","virginica")) lookup2 = structure(c("Sepal.Length","Sepal.Width","Petal.Length","Petal.Width"))

```

Sidebar {.sidebar}

```{r} tags$br()

tags$h4("Customize Scatter Plot and Clustering Settings")

tags$p("Select the variables for the X and Y axes to visualize relationships in the dataset.")

selectInput( 'xcol', label = 'X-axis variable (e.g., Sepal Length):', choices = lookup2, selected = "Sepal.Length" ) selectInput( 'ycol', label = 'Y-axis variable (e.g., Sepal Width):', choices = lookup2, selected = "Sepal.Width" )

tags$hr() tags$p("Click the barplots tab to see different bar plots. You can also visualize K-means for different values of K by going to K-means tab")

tags$p("Use the slider below to select the number of clusters (k) for k-means clustering. The data points will be grouped into k clusters.")

sliderInput( "k", "Number of Clusters (k):", min = 2, max = 10, value = 3, step = 1
) tags$p("Hover over the data points to know their values.") tags$br() tags$p("Author: ", tags$span("Ishan Gupta", style = "color: red;"))

```

Scatter Plot

row {data-height=800}

Scatter Plot

```{r}

Here, we draw the scatter plot using ggplotly

output$scatter <- renderPlotly({

# Dynamically select X and Y columns for ggplot X <- input$xcol Y <- input$ycol

# Create the ggplot scatter plot using aes() for dynamic variables p <- ggplot(df, aesstring(x = X, y = Y , col = df$Species)) + geompoint() + labs(x = X, y = Y) + theme_minimal()

# Convert ggplot object to Plotly interactive plot ggplotly(p) })

plotlyOutput('scatter', width = "100%")

Row {data-height = 200} ----------------------------------------------------------------------- ### Species(Quantile){r} output$stats <- renderTable({ df %>% groupby(Species) %>% summarise( MeanX = mean(.data[[input$xcol]]), MedianX = median(.data[[input$xcol]]), SDX = sd(.data[[input$xcol]]), MeanY = mean(.data[[input$ycol]]), MedianY = median(.data[[input$ycol]]), SD_Y = sd(.data[[input$ycol]]) ) }) tableOutput('stats')

``` Bar Plots ====================================================================== Row -----------------------------------------------------------------------

Sepal Length

```{r} p <- ggplot(df, aes(x = Species, y = Sepal.Length, fill = Species)) + geomboxplot() + thememinimal()

ggplotly(p) ```

Sepal Width

```{r} p <- ggplot(df, aes(x = Species, y = Sepal.Width, fill = Species)) + geomboxplot() + thememinimal()

ggplotly(p) ```

Row

Petal Length

```{r} p <- ggplot(df, aes(x = Species, y = Petal.Length, fill = Species)) + geomboxplot() + thememinimal()

ggplotly(p) ```

Petal Width

```{r} p <- ggplot(df, aes(x = Species, y = Petal.Width, fill = Species)) + geomboxplot() + thememinimal()

ggplotly(p)

```

K-Means{data-orientation=rows}

Row {data-height=700}

```{r}

Reactive clustering

output$clusterPlot <- renderPlotly({ # Perform K-means clustering k <- input$k xcol <- input$xcol ycol <- input$ycol

irisdata <- iris[, 1:4] kmeansresult <- kmeans(irisdata, centers = k, nstart = 25) iris$Cluster <- as.factor(kmeansresult$cluster)

# Create ggplot scatter plot plot <- ggplot(iris, aesstring(x = xcol, y = ycol, color = "Cluster")) + geompoint(size = 3, alpha = 0.7) + theme_minimal() + labs(title = paste("K-means Clustering (k =", k, ")"), x = xcol, y = ycol)

ggplotly(plot) })

Display the plot

plotlyOutput("clusterPlot")

```

Row

Cluster Summary

```{r} output$summary <- renderUI({ k <- input$k

irisdata <- iris[, 1:4] kmeansresult <- kmeans(irisdata, centers = k, nstart = 25) iris$Cluster <- as.factor(kmeansresult$cluster)

# Confusion matrix (True labels vs Predicted labels) confusion <- table(TrueLabel = iris$Species, PredictedCluster = iris$Cluster)

# Generate and style the table using kableExtra tablehtml <- knitr::kable(confusion, format = "html", table.attr = "class='table table-bordered table-hover'") %>% kableExtra::kablestyling( bootstrapoptions = c("striped", "hover", "condensed"), fullwidth = FALSE, # Use full width for the table position = "center", # Center the table fontsize = 18 # Increase font size for better readability ) %>% kableExtra::columnspec(1, color = "white", background = "#007bff") %>% # Customize the first column kableExtra::column_spec(2:ncol(confusion), background = "#f8f9fa", color = "black") # Customize other columns

# Return the HTML output correctly HTML(table_html) # Use HTML() to render the table as HTML })

In the UI part, use uiOutput to render the summary

uiOutput("summary")

```