Important Machine Learning terminology | Labels, Features, Labeled Example, Unlabeled Example

Label – Value we are trying to predict (output)
It is the variable ‘y’ in the basic linear regression
In this example, the label refers to the target or output variable that we want the machine learning model to predict. In sentiment analysis, the label can be binary, with "positive" or "negative" as the possible values. The label indicates the sentiment associated with a movie review.

Feature: Features are the input variables or attributes used by the machine learning model to make predictions. In sentiment analysis, the features can include various aspects of the movie review, such as the words used, the length of the review, or the presence of certain keywords. These features capture information that the model will use to infer the sentiment of the review.

Labeled Example: A labeled example consists of a set of features along with their corresponding labels. In our sentiment analysis example, a labeled example could be a movie review along with its associated sentiment label. For instance:

Feature: "I absolutely loved this movie! The acting was incredible, and the plot kept me engaged throughout."
Label: Positive

This labeled example provides both the input (the movie review) and the desired output (the sentiment label), which can be used to train a supervised learning model.

Unlabeled Example: An unlabeled example consists only of the input features without the corresponding labels. In sentiment analysis, an unlabeled example would be a movie review without any sentiment label. These unlabeled examples are commonly used in unsupervised learning or for making predictions on new, unseen data where the labels are not available.

For example:

Feature: "The cinematography in this film was visually stunning."