Please enable JavaScript.
Coggle requires JavaScript to display documents.
DATA REPROCESSING AND ANALYSIS image, Made by: Julieth Paola Gómez…
DATA REPROCESSING AND ANALYSIS
Processing something that has already been processed previously
Because?
Improve data quality
Help with interpretation and understanding of results
Avoid errors due to missing data, Inadequate scales, extreme values
Increases the efficiency of algorithms
Common problems in real-world data sets
Missing or null values
Noise
Different scales
Duplicates
Repeated records
Variables with widely varying ranges
Measurement errors
Data not recorded, data entry errors
Typical steps
Data Cleaning
Remove duplicates
Correct data type or typing errors
Remove noise, detect and treat outliers
Normalization
Data transformations if necessary
Scale variables that are in comparable ranges
Feature selection
Remove redundant variables
Identify which variables are useful
Algorithms, statistical criteria
Made by: Julieth Paola Gómez Sanguino
