Data Science Cheat Sheet

Data Science Pipeline

The basic process for conducting a Data Science project consists of the following steps:

1. Data Collection: Gathering raw data from various sources.
2. Data Preprocessing: Cleaning data (Data Cleaning), handling Missing Values, and removing Outliers.
3. Exploratory Data Analysis (EDA): Initial data analysis to understand the characteristics of the data.
4. Feature Engineering: Creating or transforming variables to make them suitable for Machine Learning.
5. Modeling: Training models with appropriate algorithms.
6. Evaluation: Evaluating model performance.
7. Deployment: Deploying models for real-world usage on Production systems.

Exploratory Data Analysis (EDA)

Performing EDA is an important step in discovering Insights and patterns in data.

• Descriptive Statistics: Using statistical measures such as Mean, Median, Mode, Variance, and Standard Deviation.
• Data Visualization: Creating charts such as Histograms, Scatter Plots, Box Plots, and Correlation Matrices to observe relationships between variables.

Feature Engineering

The process of improving features to increase model efficiency.

• Imputation: Filling Missing Values with Mean, Median, or other methods.
• Encoding: Converting Categorical Data to Numerical Data through techniques such as One-Hot Encoding or Label Encoding.
• Scaling: Adjusting data scales like Standardization (Z-score) and Normalization (Min-Max Scaling).
• Feature Selection: Selecting only features that are important for prediction to reduce Dimensionality.