First phase in ML project to frame the issue being addressed?

First phase in ML project to frame the issue being addressed?

• A. Definition of problem or task
• B. Model selection
• C. Data collection and preprocessing
• D. Model training

Answer: (A)

Framing the issue means clearly defining what problem you’re solving, what a successful outcome looks like, and the constraints around the project. This step sets the objective and determines how you’ll evaluate success, which in turn guides whether you’re building a classifier, a regressor, or another type of model. By articulating the goal, you also establish the data needs and the metrics that will judge performance, so every later decision—data collection, preprocessing, model choice, and training—aligns with that goal. Without a clear problem definition, efforts can drift toward the wrong objective or inadequate evaluation. Once the problem is defined, you specify what data is required and how it should be prepared, then select an appropriate model approach based on the task and data characteristics, and finally train and validate the model.

Framing the issue means clearly defining what problem you’re solving, what a successful outcome looks like, and the constraints around the project. This step sets the objective and determines how you’ll evaluate success, which in turn guides whether you’re building a classifier, a regressor, or another type of model. By articulating the goal, you also establish the data needs and the metrics that will judge performance, so every later decision—data collection, preprocessing, model choice, and training—aligns with that goal. Without a clear problem definition, efforts can drift toward the wrong objective or inadequate evaluation.

Once the problem is defined, you specify what data is required and how it should be prepared, then select an appropriate model approach based on the task and data characteristics, and finally train and validate the model.