Internship
Student login
← Data Science learning path
Data Science · Module 4 of 10

Data cleaning and wrangling

Turn messy data into analysis-ready data.

📖 Key ideas

  1. Handle missing values: drop, fill (mean/median/mode) or flag.
  2. Remove duplicates and fix data types.
  3. Standardise text: case, spaces, spellings.
  4. Detect outliers with IQR or z-scores.
  5. Reshape with merge, groupby, pivot and melt.
💡 Real-life example: Data cleaning is sorting dal from stones before cooking.
ImputationFilling missing values
OutlierUnusually extreme value
MergeJoining tables
Tidy dataEach variable a column, each row an observation