Feature Engineering & Pre-Processing
This technique is the process of replacing missing data with substituted values
What is Imputation
Models with high BLANK pay a lot of attention to training data and do not generalize on data they haven't seen before
What is Variance
This function calculates a “weighted sum” of its input, adds a bias and then decides whether a node should be “fired” or not
What is an activation function
An error that occurs when the null hypothesis is true, but is rejected
What is a Type 1 error
This is a parameter whose value is set before the learning process begins
What is a hyperparameter
This is the process by which categorical variables are converted into a form that could be provided to ML algorithms
What is One-Hot Encoding
An algorithm that takes unlabeled points and gradually learn how to cluster them into groups by computing the mean of the distance between different points.
What is K-means
This algorithm minimizes the cost function in order to minimize an error.
What is gradient descent
This visual plots the distribution of a numeric variable’s values as a series of bars
What is a Histogram
This occurs when a model learns the details and noise in the training data to the degree that it adversely impacts the execution of the model on new information
What is overfitting
This is a scaling technique in which values are shifted and rescaled so that they end up ranging between 0 and 1
What is Normalization
A graphical representation of the contrast between the true positive rate and the false positive rate at various thresholds
What is an ROC curve
This technique is used to improve the performance of a network. It revises the error and updates the weights to reduce this error
What is backpropagation
If P(A∩B) = P(A) · P(B), then events A and B are BLANK
What is Independent
This function is used when you need to come up with a range of continuous random variables
What is a Probability Density Function(PDF)
This is a data transformation method in which it replaces each variable x with a log(x)
What is Log-transform
This type of model can efficiently perform a non-linear classification using what is called the kernel trick, implicitly mapping their inputs into high-dimensional feature spaces
What is Support Vector Machine (SVM)
This activation function gives an output of X if X is positive and zeroes otherwise. It is often used for hidden layers.
What is ReLU
If the BLANK is more than then the BLANK, then we fail to reject the H0
What is the p-value and critical value
The decrease in entropy after a dataset is split on an attribute.
What is information gain
This technique groups continuous data into a smaller number of groups
What is Binning
A technique that is used to discourage the complexity of a model. It does this by penalizing the loss function.
What is regularization
This technique is used to reduce the spatial dimensions of a CNN
What is pooling
A measure that indicates the extent to which two random variables change in cycle
What is Covariance
This is an iterative technique which adjusts the weight of an observation based on the last classification
What is boosting