The Machine learning first step after collect the data, make a cleaning process in several types before make the pre-processing and apply different algorithms to make a decision what is the optimal in term of the accuracy and privacy
Now, As a Weather data before we want to use the machine learning process to optimize (such as the pollution challenge)
what is the data that will be cleaned before we can apply the algorithms?
I don't think privacy is a factor while processing your data or optimizing your machine learning algorithm's performance.
In general, you should replace missing data with NaN, smooth your data to have less fluctuations, then normalize it; before you can start using machine learning.
These steps are generally followed in the mentioned sequence irrespective of type of dataset.
Ok Mr. Utkarsh Singh
Thank you very much for this valuable answer
I wish you all the best, successes and good luck
- Badges
- Science topic