In Machine learning, Data preprocessing techniques are involved in the process of cleaning the data and making it more suitable for building the machine learning model. The main goal is to increase the accuracy and efficiency of the model.

The following steps are data preprocessing techniques:
 •  Data cleaning: It removes irrelevant data, noisy data, and inaccurate data in the sample. Binning, Regression, and clustering are useful methods for data cleaning. The main goal is to provide a simple, complete, and clear sample set for machine learning.
 •  Data integration: The process of combining multiple sources into a single sample.
 •  Data Reduction: It refers that the process of reducing the data in the sample. Dimensionality reduction methods are used to reduce the number of features/attributes in the sample and provide the desired features for the sample.
 •  Data Transformation: It is the final step in the data preprocessing technique that refers the transforming all data into the same unit after all preprocessing methods. Smoothing, Aggregation, Discretization, generalization, and normalization are the methods used to transform the data.