plot_outlier.tbl_dbi: Plot outlier information of numerical data diagnosis in the...
In bit2r/kodlookr: Korean Help Resources for the dlookr Package

Description Usage Arguments Details Outlier diagnostic information See Also Examples

The plot_outlier() visualize outlier information for diagnosing the quality of the numerical(INTEGER, NUMBER, etc.) column of the DBMS table through tbl_dbi.

## S3 method for class 'tbl_dbi'
plot_outlier(
  .data,
  ...,
  col = "steelblue",
  in_database = FALSE,
  collect_size = Inf,
  typographic = TRUE
)

`.data`	a tbl_dbi.
`...`	one or more unquoted expressions separated by commas. You can treat variable names like they are positions. Positive values select variables; negative values to drop variables. If the first expression is negative, plot_outlier() will automatically start with all variables. These arguments are automatically quoted and evaluated in a context where column names represent column positions. They support unquoting and splicing.
`col`	a color to be used to fill the bars. The default is "lightblue".
`in_database`	Specifies whether to perform in-database operations. If TRUE, most operations are performed in the DBMS. if FALSE, table data is taken in R and operated in-memory. Not yet supported in_database = TRUE.
`collect_size`	a integer. The number of data samples from the DBMS to R. Applies only if in_database = FALSE.
`typographic`	logical. Whether to apply focuses on typographic elements to ggplot2 visualization. The default is TRUE. if TRUE provides a base theme that focuses on typographic elements using hrbrthemes package.

The scope of the diagnosis is the provide a outlier information. Since the plot is drawn for each variable, if you specify more than one variable in the ... argument, the specified number of plots are drawn.

The plot derived from the numerical data diagnosis is as follows.

With outliers box plot
Without outliers box plot
With outliers histogram
Without outliers histogram

See vignette("diagonosis") for an introduction to these concepts.

plot_outlier.data.frame, diagnose_outlier.tbl_dbi.

library(dplyr)

# connect DBMS
con_sqlite <- DBI::dbConnect(RSQLite::SQLite(), ":memory:")

# copy heartfailure to the DBMS with a table named TB_HEARTFAILURE
copy_to(con_sqlite, heartfailure, name = "TB_HEARTFAILURE", overwrite = TRUE)

# Using pipes ---------------------------------
# Visualization of all numerical variables
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier()
  
# Positive values select variables
 con_sqlite %>% 
   tbl("TB_HEARTFAILURE") %>% 
   plot_outlier(platelets, sodium)
  
# Negative values to drop variables, and In-memory mode and collect size is 200
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier(-platelets, -sodium, collect_size = 200)
  
# Positions values select variables
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier(6)
  
# Positions values select variables
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier(-1, -5)
  
# Not allow the typographic elements
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier(-1, -5, typographic = FALSE)

# Using pipes & dplyr -------------------------
# Visualization of numerical variables with a ratio of
# outliers greater than 1%
con_sqlite %>% 
  tbl("TB_HEARTFAILURE") %>% 
  plot_outlier(con_sqlite %>% 
                 tbl("TB_HEARTFAILURE") %>% 
                 diagnose_outlier() %>%
                 filter(outliers_ratio > 1) %>%
                 select(variables) %>%
                 pull())

# Disconnect DBMS   
DBI::dbDisconnect(con_sqlite)