For Homework 4, I am practicing using plotly to create some interactive graphs, based on the instacart dataset from p8105.datasets. Instacart is a same-day grocery delivery service. This dataset includes information such as product number and name, the aisle and department, and the day of the week and hour of the day on which the order was placed.
First, I created a smaller dataset to work with, as this dataset is quite large. I limited it to items only in the frozen foods aisle, resulting in 100426 observations.
The graphs show a few interesting details. For example, we can see in Chart A, that the median hour at which individuals place their order (for items from the frozen food department), across all aisles, is at either 1pm or 2pm. We can also see, in Chart B, that the aisles with the most orders in the department are ice cream and frozen produce.
---
title: "Dashboard"
output:
flexdashboard::flex_dashboard:
orientation: columns
vertical_layout: fill
source: embed
---
For Homework 4, I am practicing using `plotly` to create some interactive graphs, based on the `instacart` dataset from `p8105.datasets`. Instacart is a same-day grocery delivery service. This dataset includes information such as product number and name, the aisle and department, and the day of the week and hour of the day on which the order was placed.
```{r setup, include=FALSE}
library(flexdashboard)
library(tidyverse)
library(p8105.datasets)
library(plotly)
```
```{r clean_data, message = FALSE, warning = FALSE}
n_items_df =
instacart %>%
group_by(order_id) %>%
summarize(n_items = n())
n_frozen_df =
instacart %>%
filter(department == "frozen") %>%
group_by(order_id) %>%
summarize(n_frozen = n())
instacart_df =
instacart %>%
left_join(n_items_df) %>%
left_join(n_frozen_df) %>%
select(order_id, n_items, n_frozen, product_name, reordered, order_dow, order_hour_of_day, aisle, department) %>%
filter(department == "frozen")
```
First, I created a smaller dataset to work with, as this dataset is quite large. I limited it to items only in the frozen foods aisle, resulting in `r nrow(instacart_df)` observations.
The graphs show a few interesting details. For example, we can see in Chart A, that the median hour at which individuals place their order (for items from the frozen food department), across all aisles, is at either 1pm or 2pm. We can also see, in Chart B, that the aisles with the most orders in the department are ice cream and frozen produce.
Column {data-width=650}
-----------------------------------------------------------------------
### Chart A
```{r box_plot}
instacart_df %>%
mutate(aisle = fct_reorder(aisle, order_hour_of_day)) %>%
plot_ly(
x = ~order_hour_of_day, y = ~aisle, color = ~aisle,
type = "box", colors = "viridis") %>%
layout(
title = "Distribution of Order Times by Aisle",
xaxis = list(
title = "Hour",
ticktext = list("6am","12pm","6pm", "12am"),
tickvals = list(6, 12, 18,24)
),
yaxis = list(title = "Aisle")
)
```
Column {data-width=350}
-----------------------------------------------------------------------
### Chart B
```{r bar_graph}
instacart_df %>%
count(aisle) %>%
mutate(aisle = fct_reorder(aisle, n)) %>%
plot_ly(
x = ~aisle, y = ~n, color = ~aisle,
type = "bar", colors = "viridis"
) %>%
layout(
title = "Frozen Food Orders by Aisle",
xaxis = list(title = "Aisle"),
yaxis = list(title = "Number of Items")
)
```
### Chart C
```{r scatter_plot}
instacart_df %>%
mutate(text_label = str_c(n_frozen, " Frozen, ", n_items, " Total Items, ", "Ratio: ", n_frozen/n_items)) %>%
plot_ly(
x = ~n_frozen, y = ~n_items,
alpha = .5, type = "scatter", mode = "markers", colors = "viridis", text = ~text_label) %>%
layout(
title = "Proportion of Frozen Food Items to Total Order Size",
xaxis = list(title = "Number of Frozen Items Ordered"),
yaxis = list(title = "Number of Items Ordered")
)
```