I need help with the assignment of defining a Machine Learning example with some questions.

The questions are based on the Kernel Machine concept (SVM - Support Vector Machine)

Pre-requisite knowledge of k-NN Algorithm, Bayes Classifier, Gradiant Distribution, Linear Regression, Logistic Regression, Decision Trees, CART Algorithm, Loss Functions (Squared, Log, Absolute, Exponential, Zero-One, and Hinge), and Regularizations (L2 Regularizor, L1 Regularizor, Lp Norm etc)

The assignment goes as below:

Instruction: Map your own problem taking the ERM approach. You will specifically describe the particulars of the problem you would like to solve.

Q1: What is the problem space/context?

Q2: What do you want to learn?

Q3: What are the constraints/limitations of your data?

Q4: What algorithm, loss function, and regularizer will you use? Why?