Data Pre-processing for Machine Learning
The success of machine learning algorithms depend on the quality of the data. If the dataset is not in the desirable representation the machine learning model finds difficulty in extracting knowledge from it, hence the need for pre-processing the data before applying it into machine learning models. Data pre-processing has noticeable effects on the following:
- Accuracy.
- Training time.
- Generalization.
- Model’s fairness. Data pre-processing includes data cleaning, normalization, transformation, feature extraction and selection. In the following section we discuss some data pre-processing techniques:
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.