hipster-veggies

Find Select Trending Vegetables from Twitter 🥕

Analysis

Going to use the rtweet library and some assorted tidyverse things.

library(rtweet) # devtools::install_github("mkearney/rtweet")
library(tidyverse)

# I persisted these credientials in my ~/.Rprofile
twitter_token <- create_token(
app = Sys.getenv("appname"),
consumer_key = Sys.getenv("twitter_key"),
consumer_secret = Sys.getenv("twitter_secret"))

Searching for the top most recent tweets for each vegetable from the article: 7 Hipster Vegetables Most Likely to Dethrone Kale. The documentation for the query arguement to search for more words by using the 'OR' operator did not work, so multiple searches were ran.

radish_tweets <- search_tweets(q="radish", 
                               n = 1000, 
                               include_rts = FALSE, retryonratelimit = TRUE, lang = "en") 

cauliflower_tweets <- search_tweets(q="cauliflower", 
                                    n = 1000, 
                                    include_rts = FALSE, retryonratelimit = TRUE, lang = "en")

turnip_tweets <- search_tweets(q="turnip", 
                                    n = 1000, 
                                    include_rts = FALSE, retryonratelimit = TRUE, lang = "en")

jimica_tweets <- search_tweets(q="jicama", 
                                    n = 1000, 
                                    include_rts = FALSE, retryonratelimit = TRUE, lang = "en")

rc_tweets <- search_tweets(q="rainbow chard", 
                               n = 1000, 
                               include_rts = FALSE, retryonratelimit = TRUE, lang = "en") 


bs_tweets <- search_tweets(q="brussels sprout", 
                                    n = 1000, 
                                    include_rts = FALSE, retryonratelimit = TRUE, lang = "en") 


ks_tweets <- search_tweets(q="kabocha squash", 
                                    n = 1000, 
                                    include_rts = FALSE, retryonratelimit = TRUE, lang = "en")

#ks_tweets <- dplyr::filter(ks_tweets, lang == "en")

A preview of tweets about rainbow chard.

head(rc_tweets$text)
#> [1] "Vegan Caes sitch Rainbow chard Brussels and Napa cabbage w/ pine nut Parm recipe adopted from @thefirstmess https://t.co/YG7Ca264wC https://t.co/5cDqg1VRpR"                                                                                                                               
#> [2] "Vegan caes sitch with rainbow chard Napa cabbage Brussels and smoky chic peas a la @thefirstmess #pinenutparm #yesbread #yesveggies https://t.co/r55ayJgUy9 https://t.co/wqGBKDl9uG"                                                                                                       
#> [3] "#comfort grilled tri tip & eggplant w/ arugula chimichurri sauce & happy boy farms rainbow chard!! john_dickman _hawko @marciadorsey https://t.co/3v0clvqDUt"                                                                                                                      
#> [4] "Did you know that rainbow chard is a mix of chard varieties, not just one plant? Visit the market today for fresh, organically grown rainbow chard from Ground Stew Farms. \U0001f308 https://t.co/O3rAQ0m0i0"                                                                             
#> [5] "#photobomb by rainbow Swiss chard! Blue curly kale growing with chard,  basil & more... all… https://t.co/MwM4xeaENo"                                                                                                                                                                  
#> [6] "@JBGorganic will have tomatoes, fresh lettuce, beet bunches, broccoli, green cabbage, napa cabbage, carrot bunches, fennel, a bouquet of radishes, sweet potatoes, butternut squash, turnips, cilantro, dill, parsley, arugula, bok choy, braising mixed greens, rainbow chard & more!"

Let's combine these data sets into one for graphing while preserving the original vegetable type by creating an additional column.

bs_tweets$vegetable <- "brussels sprout"
cauliflower_tweets$vegetable <- "cauliflower"
jimica_tweets$vegetable <- "jimica"
ks_tweets$vegetable <- "kabocha squash"
radish_tweets$vegetable <- "radish"
rc_tweets$vegetable <- "rainbow chard"
turnip_tweets$vegetable <- "turnip"

all_veggies <- rbind(bs_tweets, cauliflower_tweets, jimica_tweets, ks_tweets, radish_tweets, rc_tweets, turnip_tweets)

Let's plot the data to observe any trends.

gg_veggies <- all_veggies %>%  group_by(created_at, vegetable) %>% 
                  summarise(n = n() ) %>% 
            ggplot(., aes(x = created_at)) +
                geom_freqpoly(aes(color = vegetable)) +
                scale_color_brewer(palette = "Set1") +
                theme_minimal() +
                theme(plot.title = element_text(face = "bold"), legend.title = element_blank()) +
                labs( x = NULL, y = "count (log scale)", title = "Frequency of Tweets Mentioning Hipster Vegetables", subtitle = "from the past 9 days", caption = "\nSource: Data collected from Twitter's REST API via rtweet"
  ) + scale_y_log10(breaks = c(1, 10, 100, 200, 300, 400))

gg_veggies

fin 🎉

Extra

Here are some other articles about Hipster Vegetables:

1. The 22 Most Hipster Foods On The Planet - HuffPost

2. What Will Be the Next Hipster Vegetable?

Contact Me

Jasmine Daly (@jasdumas) twitter, jasdumas.github.io