What is Data Preprocessing

Data preprocessing is the process of cleaning and preparing raw data before feeding it into a machine learning model. Real-world data is often messy, with missing values, wrong formats, or irrelevant information, and models cannot learn properly from data like that.

Why preprocessing matters

• Raw data often has errors, duplicates, or missing values.
• Machine learning models need numerical, structured input to work correctly.
• Clean data directly improves model accuracy and performance.

A simple preprocessing example

Here, we load a dataset, check for missing values, and remove duplicate rows, three of the most common preprocessing steps.

import pandas as pd

df = pd.read_csv("data.csv")

print(df.isnull().sum())

df = df.drop_duplicates()

print(df.shape)

df.isnull().sum() checks how many missing values exist in each column, so you know what needs fixing. df.drop_duplicates() removes repeated rows, since duplicates can bias the model. df.shape then shows the dataset's updated size after cleaning.

Preprocessing is usually the first real step in any machine learning project, done after collecting data and before building any model.