Data Mining and Machine Learning Concepts

These flashcards cover key concepts and terminology from the Data Mining and Machine Learning lecture notes.

The first phase of the Data Mining Process is __.

Business Understanding

In the Data Mining Process, __ is the phase where data is collected and described.

Data Understanding

The process of cleaning and transforming data into the required format is known as __.

Data Preparation

The variable to predict in data mining is referred to as a __.

Label

Input variables used to predict outcomes are called __.

Features

An example of a __ feature is age or income.

Numerical

Features that contain non-numerical values and are grouped into categories are known as __ features.

Categorical

In __ learning, models learn from labeled data to make predictions.

Supervised

The type of learning where the model learns from unlabeled data to identify patterns is called __ learning.

Unsupervised

The function used in supervised learning to quantify prediction accuracy is called __ function.

Loss

The process of adjusting model parameters to minimize the loss function is termed __.

Optimization

The method for handling outliers can include __ or keeping them based on context.

Removing

One-hot encoding is used to transform __ features into numerical values.

Categorical

Normalization is important for features with __ ranges to ensure balanced comparisons.

Large

The purpose of the __ method is to minimize redundancy by removing irrelevant features.

Feature Selection

In imbalanced datasets, __ involves randomly removing samples from the majority class.

Downsampling

The __ is a table that describes the performance of a classification model.

Confusion Matrix

The ratio of true positive predictions to all predicted positives is called __.

Precision

The probability that measures the chance of a positive instance being predicted correctly is called __.

Recall

A __ curve plots the True Positive Rate against the False Positive Rate.

ROC (Receiver Operating Characteristic)

The harmonic mean of precision and recall is known as the __ score.

F1

Decision trees are used for both classification and __ tasks.

Regression

Binary classification involves distinguishing between __ classes.

Two

The ultimate goal of supervised learning is to make __ that generalize well to unseen data.

Predictions

The algorithms in supervised learning use __ data tagged with correct outputs.

Labeled

A technique that adds a penalty term to the loss function to prevent overfitting is called __.

Regularization

In logistic regression, the mapping function used to predict probabilities is termed __.

Sigmoid Function

The learning mechanism of a neural network involves weights assigned to connections which determine __.

Influence

Deep Neural Networks consist of multiple layers, including __, hidden, and output layers.

Input

A __ model uses probabilities to predict class membership based on input features.

Naïve Bayes

Conditional independence states that events A and B are independent given __.

C

AUC stands for __, which measures the performance of a classification model.

Area Under the ROC Curve

Features with independent variables in Naïve Bayes are assumed to be __.

Conditionally independent.