Handling missing data (imputation techniques)

This chapter has primarily focused on supervised algorithms that can be applied to solve various problems. Another important concept that was briefly discussed in Chapter 1 is handling missing data. The basic types of dealing with missing data were previously discussed. Below are various common approaches to missing data:
  • 1) Simply remove any samples with N/A values: This is the simplest method of treating missing data, and the disadvantage of this approach is that the number of samples could be drastically reduced due to the removal of missing info. For example, if 1000 out of 2000 wells miss one important feature, removing wells with missing features will result in only 1000 wells. This is 50% less ...

Get Machine Learning Guide for Oil and Gas Using Python now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.