This blog post will analyze water resources in Africa. This is an interesting post, because it makes maps by using the ggmap package, and also it exposes some potential problems the tidymodels package has, especially when data has missing values. library(tidyverse) library(lubridate) library(tidymodels) library(widyr) library(ggthemes) library(ggmap) library(countrycode) theme_set(theme_bw()) water <- read_csv('https://raw. Read more

This blog post follows up Julia Silge’s penguin blog post (link). In her post, she only uses the bootstrap method, but here in this data analysis, I will use both bootstrap and 10-fold cross validation for the purposes of practicing the tidymodels meta-package. Read more

This blog post is diving into an interesting dataset about Netflix shows. Some graphs (network plots) will be shown, and since the dataset contains text information, using LASSO model to predict some show rating is suitable. library(tidyverse) library(lubridate) library(widyr) library(tidylo) library(tidytext) library(ggraph) library(glmnet) library(broom) theme_set(theme_bw()) netflix <- read_csv('https://raw. Read more

In this blog post, I will analyze the post offices in the U.S. in terms of when they were established and when they were discontinued. library(tidyverse) library(geofacet) library(tidytext) theme_set(theme_bw()) post_offices <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-04-13/post_offices.csv') %>% filter(longitude < 100, established > 1000, established < discontinued, discontinued < 2020) %>% select(-contains("gnis"), -alt_name, -orig_name) %>% mutate(name = str_to_title(name), est_decade = 10 * (established %/% 10), dis_decade = 10 * (discontinued %/% 10)) post_offices ## # A tibble: 82,173 x 17 ## name state county1 county2 county3 orig_county established discontinued ## <chr> <chr> <chr> <lgl> <lgl> <chr> <dbl> <dbl> ## 1 Aaron MO BATES NA NA Bates 1895 1933 ## 2 Aaron GA BULLOCH NA NA Bulloch 1909 1920 ## 3 Aaron SC ANDERSON NA NA Anderson 1892 1899 ## 4 Abadyl MO CHRISTI~ NA NA Christian 1895 1919 ## 5 Abattis MO WARREN NA NA Warren 1878 1904 ## 6 Abaugh AR NEWTON NA NA Newton 1928 1954 ## 7 Abba GA IRWIN NA NA Irwin 1884 1954 ## 8 Abbey AL FRANKLIN NA NA Franklin 1898 1902 ## 9 Abbie Joe LA BEAUREG~ NA NA Beauregard 1918 1925 ## 10 Abbot ME PISCATA~ NA NA Piscataqui~ 1825 1912 ## # . Read more

Deforestation is a contemporary topic nowadays, and in this blog post, I will analyze a few datasets from TidyTuesday to shed some light in this regard. library(tidyverse) library(scales) library(worlddatajoin) theme_set(theme_bw()) forest <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-04-06/forest.csv') forest_area <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-04-06/forest_area.csv') brazil_loss <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-04-06/brazil_loss.csv') soybean_use <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-04-06/soybean_use.csv') vegetable_oil <- read_csv('https://raw. Read more

In this blog post, I will carry a slew of interesting data visualizations on the votes in the U.N. The datasets are from TidyTuesday. Also, this is my first time using my own package worlddatajoin. You can download it from my Github by typing devtools::install_github(" Read more

This blog post analyzes vidoe games time-series data from TidyTuesday. I personally never play video games and know nothing about them, but through data analysis in this post, I can use data to shed some light on what video games are popular over time and which games gained popularity during the COVID lockdown (March 2020). Read more

This blog post analyzes the Bechdel test dataset from TidyTuesday for a wide range of movies. library(tidyverse) library(lubridate) library(scales) library(tidytext) library(glmnet) library(Matrix) library(tidymodels) library(vip) theme_set(theme_bw()) raw_bechdel <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-03-09/raw_bechdel.csv') %>% rename(imdb = imdb_id) %>% mutate(decade = 10 * (year %/%10)) movies <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-03-09/movies.csv') %>% mutate(imdb = str_remove(imdb, "tt")) Decade-wise bechdel ratings: Read more

This blog post analyzes the ads for the Super Bowl. This is an intersting dataset coming from TidyTuesday. In this post, I will visualize the data from various perspectives, and use machine learning model to analyze it with the tidymodels meta-package used. Load the packages: Read more

In this blog post, I will analyze employment data from TidyTuesday, and I will also use my own R package ggDoubleHeat to make some visualization. Load the packages: library(tidyverse) library(scales) library(ggDoubleHeat) theme_set(theme_bw()) Data loading and initial cleaning: employed <- read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-02-23/employed.csv') %>% mutate(minor_occupation = str_remove_all(minor_occupation, "\\-")) %>% add_count(industry) %>% filter(n > 100, ! Read more