What is the difference between labeled data and data used for classification prediction?

Question

I feel very confused when attempting to handle data, especially for classification purposes. I have a dataset that initially lacked labels, so I applied a labeling method to it. Afterwards, I used the features (continuous variable) I had labeled to make predictions for labeling.

I have two questions:

1. Is this the correct approach for handling data in a classification task?
2. If data can be labeled using a method, why should I resort to using machine learning for classification? Are there additional benefits to using machine learning? example

Answer 1

1. There is no explanation of the "labeling method" you talked about so I couldn't provide my opinion here.

2. If data can be labeled using methods, machine learning is not needed. But as you are studying right now, that is not possible to let you start with dataset will billions of complicated data, for example image data or audio data.

There is no benefit to uses machine learning on data that you can label it using methods, algorithms as it is not worth the time to train a model for it. Machine learning is generally used for complicated data such as images, example recognize given images data.