healthcareai (version 1.2.4)

featureAvailabilityProfiler: Calculate and plot data availability over time

Description

Helps you determine how much data is present in each feature, by hour, after a particular event (like patient admit)

Usage

featureAvailabilityProfiler(df, startDateColumn = "AdmitDTS",
  lastLoadDateColumn = "LastLoadDTS", plotProfiler = TRUE)

Arguments

df

A dataframe

startDateColumn

Optional string of the column name, representing the date of the starting event of interest (e.g., patient admit)

lastLoadDateColumn

Optional string of the column name, representing the date the row was loaded into the final dataset (i.e., via daily ETL)

plotProfiler

Default is TRUE. Whether to plot profiler results

Value

a list, that has as many vectors as columns in the original dataframe, with each vector holding the percentage full for each hour

References

http://healthcareai-r.readthedocs.io

See Also

healthcareai

Examples

# NOT RUN {
df <- data.frame(a = c(2,1,3,5,4,NA,7,NA),
                 b = c(0.7,-2,NA,-4,-5,-6,NA,NA),
                 c = c(100,300,200,NA,NA,NA,NA,500),
                 d = c(407,500,506,504,NA,NA,NA,405),
                 admit = c('2012-01-01 00:00:00','2012-01-01 00:00:00',
                           '2012-01-01 12:00:00','2012-01-01 12:00:00',
                           '2012-01-02 00:00:00','2012-01-02 00:00:00',
                           '2012-01-02 12:00:00','2012-01-02 12:00:00'),
                 loaded = c('2012-01-03 00:00:00','2012-01-03 00:00:00',
                            '2012-01-03 00:00:00','2012-01-03 00:00:00',
                            '2012-01-03 00:00:00','2012-01-03 00:00:00',
                            '2012-01-03 00:00:00','2012-01-03 00:00:00'))
str(df)
head(df)

d <- featureAvailabilityProfiler(df = df,
                                 startDateColumn = 'admit',
                                 lastLoadDateColumn = 'loaded')
d # Look at the data in the console                                
# }