Little Bits of Artificial Intelligence

1. Cartesian Coordinates (2D Plane)

• What it is: Knowing how to plot a point like (x, y) on a graph.
• Why it’s needed: SVM visualizes data points in 2D (or more), and places a line or plane to separate them.

Example: We know that (50, 20) means “income is 50, debt is 20”.

2. Linear Equations

• What it is: Equation of a straight line:
  

  y = mx + c or in SVM form: w1 * x1 + w2 * x2 + b = 0

• Why it’s needed: This is how the decision boundary is drawn in SVM.

Example: We understand that a line can be used to split points on a graph — like separating apples and oranges.

3. Basic Algebra

• What it is: Working with variables, coefficients, and solving for unknowns.
• Why it’s needed: We’ll rearrange and calculate using the SVM decision formula.

Example: Solving 2x + 3 = 7 feels easy to you.

4. Inequalities and Signs

• What it is: Understanding greater than (>) and less than (<) comparisons.
• Why it’s needed: SVM decides the class based on whether the formula output is positive or negative.

Example: If result > 0, it’s a “Good customer”; if < 0, it’s “Bad”.

6. Vectors (Basic Idea)

• What it is: A quantity with direction and magnitude, like an arrow.
• Why it’s needed: SVM uses weight vectors (w1, w2) to define the direction of the separating line or plane.

Example: Think of “Income + (-Debt)” as a direction in space.

7. Dot Product (Advanced Use Case)

• What it is: A way to measure similarity/direction between two vectors.
• Why it’s needed: Used inside SVM kernels, especially when working with non-linear data.

Example: Helpful for SVM variants but not essential for linear SVMs.