Indian Food Orders
indian-food-orders.Rmd
knitr::opts_chunk$set(fig.width=8, fig.height=5) Visualization
box <- ggplot2::ggplot(data=name_and_quant, mapping=aes(x=Meat.Veg, y=Product.Price, color=Meat.Veg)) +
scale_colour_manual(values = c("tomato3","darkgreen")) +
theme_bw() +
ggtitle("Price Distributions for Meat & Vegetable Dishes") +
xlab("Type of Dish") +
ylab("Price") +
geom_boxplot()
plotly::ggplotly(box)
#name_and_quant$
p <- name_and_quant %>%
ggplot2::ggplot( aes(Product.Price, Total.Quantity, color=Meat.Veg, label=Item.Name) ) +
scale_colour_manual(values = c("tomato3","darkgreen")) +
geom_jitter(width=0.15) +
theme_bw() +
ggtitle("Items' Number of Orders by Price") +
xlab("Item Price") +
ylab("Total Number of Orders")
plotly::ggplotly(p)Background Information
The item price is probably in pounds, since the data comes from a restaurant in London. I’m more interested in prices relative to each other than the prices themselves.
Description of data source
This dataset is from https://www.kaggle.com/henslersoftware/19560-indian-takeaway-orders. The set used has ~74000 rows containing information about an Indian restaurant’s order history from 01-09-2015 to 12-07-2019.
Each row represents one order of one item. The columns included are order number, date ordered, item name, item price, quantity of item ordered, and total products in order.
Reflections on project…
What ideas/suggestions from Claus Wilke helped shape your visualization?
Wilke’s scatterplot of birds’ bill length against their skull size– with color indicating the birds’ sex – gave me the idea to differentiate between meat/non-meat dishes with color. As a sometime vegetarian, I’m interested in comparing the price and popularity of these two groups!
Is there anything more you wish you could do with this data?
YES. I wish I could do cooler things with it.
I’m still a noob when it comes to graphics and visualizations. This source had data for a whole other restaurant, and I wish I could have compared the two venues somehow. I also wish I could have made my graph look cooler!
Also, I wish I knew what all the names of the dishes meant. I’m not super familiar with Indian food, so it’s entirely possible I missed some meat dishes because I didn’t know what words to look for.
AND FINALLY! I wish I could have gotten my tag_as_meat() function working. Because of time constraints, I ended up using repetitive code to check for each individual meat word. :’(
What were the most interesting or frustrating technical aspects of doing this?
The most interesting aspect was deciding which comparisons I wanted to make, and how I would represent them visually.
The most frustrating aspect was definitely wrangling the data. Writing an example row from the “end goal” dataset helped… but getting there took quite the struggle!