Confident Data Skills by Kirill Eremenko

Author:Kirill Eremenko
Language: eng
Format: epub
Publisher: Kogan Page Limited
Published: 2017-12-18T16:22:28+00:00

2 Fit the classifier to the training set Fitting our random forest classifier to the training set enables it to learn how to make future predictions for new data points. We can then compare these predictions against the actual results in our dataset in order to see how accurate the classifier is.

The random forest algorithm will randomly select N subsets from your dataset, where N is the number of trees you specified for the parameter in Step 1. These subsets can be overlapping; however, no two of them will be identical.

Once the subsets have been selected, each one will be used as an input dataset to build a unique classification tree. In this way, each classification tree only sees its own subset of the data and has no knowledge of the fact that the actual dataset is broader. This approach ensures variety in the generation of the trees – which is precisely where the ‘power of the crowd’ comes from in the random forest algorithm.

From this logic, to help the algorithm make more accurate predictions, we can simply add information to our dataset – the more data we have in our training set, the more accurate our algorithm’s forecast will be.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.