what is machine learning?
the use and development of computer systems with ability to learn and discover patterns in data
wha is broadcasting?
numpy feature that enables mathematical operations to be applied to arrays of different sizes
what is classification?
one of two classes of methods in supervised learning, where the label is a categorical value
what are the two types of classification?
binary and multi-class
what is a dataframe?
a data table or spreadsheet with row and column headers
what are features?
input variables which are predictive data elements
what is generalization?
a model’s ability to adapt to new, previously unseen data
what are heuristics?
a way of solving problems where the objective is to produce a solution within a reasonable time frame
what is a label?
in supervised learning, the “answer”
what is regression?
another method of supervised learning, where the label is any real valued number
what is unsupervised learning?
a class of machine learning problems in which labeled data are not available
what is artificial intelligence?
a broad, all encompassing term that captures the research and implementation of systems that are capable of performing tasks intelligently.
what is deep learning?
a subset of machine learning that uses a special type of ml model, neural networks, as its underlying algorithm
what is a data matrix?
consists of a collection of examples that are either labeled or unlabeled.
what does regression predict?
continuous numbers
what does classification predict?
predicts categorical values
what does supervised learning attempt to discover?
the relationship between features and an associated label for prediction
what does unsupervised learning attempt to discover?
patterns in data without the use of training data containing labeled examples
what is clustering?
an unsupervised learning technique that groups subsets of data that are collectively similar to one another based on the similarity of their feature values
what is a binary classification problem?
when the label we are trying to predict belongs to only two possible distinct values
what is a multi-class classification problem?
when the label we are trying to predict belongs to multiple distinct values
what is a regression problem?
when the label we are trying to predict belongs to a real number
what is cd?
change directory command
what is pwd?
present working directory