Machine Learning Decision Trees Quiz

<p>The branch is considered a terminal node. (C)</p> Signup and view all the answers

<p>Recursively applying the splitting process to create branches. (C)</p> Signup and view all the answers

<p>It adapts its performance at tasks based on previous experiences. (A)</p> Signup and view all the answers

<p>Regression (B)</p> Signup and view all the answers

<p>Positive examples that are correctly classified as positive. (D)</p> Signup and view all the answers

<p>True positive rate (D)</p> Signup and view all the answers

<p>To categorize data into groups without pre-labeled examples. (D)</p> Signup and view all the answers

<p>It requires labeled training data to learn effectively. (A)</p> Signup and view all the answers

<p>To ensure that algorithms can classify future input correctly. (A)</p> Signup and view all the answers

<p>160 (D)</p> Signup and view all the answers

<p>Machine learning seeks to improve performance through data-based experience. (B)</p> Signup and view all the answers

<p>Supervised learning (D)</p> Signup and view all the answers

<p>To achieve pure sub-datasets regarding the target feature (D)</p> Signup and view all the answers

<p>The final predictions made for new instances (A)</p> Signup and view all the answers

<p>200 (A)</p> Signup and view all the answers

<p>Semi-supervised learning (A)</p> Signup and view all the answers

<p>Both classification and regression tasks (D)</p> Signup and view all the answers

<p>To indicate when to stop splitting the dataset (C)</p> Signup and view all the answers

<p>To achieve the smallest possible WCSS (B)</p> Signup and view all the answers

<p>SSE is similar for each group (B)</p> Signup and view all the answers

<p>To partition observations into specified clusters based on proximity to cluster means (D)</p> Signup and view all the answers

<p>Markov Decision Process (MDP) (B)</p> Signup and view all the answers

<p>It can be challenging to determine the appropriate number of clusters, K (A)</p> Signup and view all the answers

<p>To explore the environment and make decisions (D)</p> Signup and view all the answers

<p>Reinforcement learning does not provide explicit guidance for actions. (C)</p> Signup and view all the answers

<p>After assigning objects to the closest cluster (C)</p> Signup and view all the answers

<p>Squared Euclidean distance (C)</p> Signup and view all the answers

<p>The optimal policy to maximize long-term rewards (A)</p> Signup and view all the answers

<p>Randomly choose an initial state (B)</p> Signup and view all the answers

<p>Clusters may be misrepresented due to differing scales in dimensions (A)</p> Signup and view all the answers

<p>To determine the number of potential clusters to form (B)</p> Signup and view all the answers

<p>The reward received after executing an action (A)</p> Signup and view all the answers

<p>When there are no changes in cluster assignments (B)</p> Signup and view all the answers

<p>Its reliance on a distance measure that cannot be applied effectively to nominal data (C)</p> Signup and view all the answers

<p>They are generally fast to train and easy to interpret. (B)</p> Signup and view all the answers

<p>Attributes are independent of each other. (B)</p> Signup and view all the answers

<p>To group similar items together. (B)</p> Signup and view all the answers

<p>Manual inspection and distance measures. (D)</p> Signup and view all the answers

<p>They are guaranteed to provide the highest accuracy. (C)</p> Signup and view all the answers

<p>Using the Euclidean distance formula. (A)</p> Signup and view all the answers

<p>High similarity within clusters and low similarity across clusters. (B)</p> Signup and view all the answers

<p>Their accuracy is often not state-of-the-art. (A)</p> Signup and view all the answers

The study of algorithms that allow computers to learn from data.

Data used to train a machine learning model.

Data used to evaluate the performance of a trained machine learning model.

Examples that are correctly classified as positive by the machine learning model.

Examples that are correctly classified as negative by the machine learning model.

Examples that are incorrectly classified as positive by the machine learning model.

Examples that are incorrectly classified as negative by the machine learning model.

A measure of the proportion of examples that are correctly classified by the machine learning model.

A machine learning technique where an algorithm learns from labeled data to predict a target variable. The goal is to find patterns and relationships between input features and output labels.

A machine learning technique where an algorithm learns from unlabeled data to discover hidden patterns and structures. The focus is on extracting insights and creating meaningful representations from the data itself.

A machine learning technique where an algorithm learns from a combination of labeled and unlabeled data. It leverages the strengths of both supervised and unsupervised learning, combining labeled data for accuracy and unlabeled data for efficiency.

A machine learning technique where an algorithm learns through trial and error, interacting with an environment to maximize rewards. The algorithm learns from its actions and continuously improves its decision-making process.

A supervised learning algorithm used for both classification and regression tasks. It creates a tree-like structure to represent decision-making processes based on input features.

A process in decision tree algorithms where the dataset is divided into subsets based on the values of the most informative feature. The goal is to create subsets with similar target feature values.

The most informative feature in a decision tree is the one that best separates the target feature values into groups with the highest purity. It minimizes the chances of misclassifying or predicting incorrectly.

Nodes in a decision tree that represent predictions made by the model. They are the final outcomes of the decision-making process.

A measure used to select the best attribute for splitting data in a decision tree, aiming to create the most effective partitions.

A recursive process that splits data based on chosen attributes, aiming to create branches leading to nodes where all instances belong to the same class.

A type of machine learning model that represents decision rules as a tree-like structure, where each internal node represents a test on an attribute, each branch represents a possible outcome, and each leaf node represents a class.

A key step in building a decision tree, where the best attribute for splitting the data is chosen based on a metric like information gain.

A decision tree can also estimate the probability that an instance belongs to a particular class by counting instances of that class in the leaf node reached by the instance.

A machine learning task where the goal is to assign data points to predefined categories or classes based on their characteristics.

A decision tree learning model, known for its speed and interpretability. It uses a tree-like structure to make predictions based on a series of decisions.

A decision tree learning model similar to ID3/C4.5. It is also fast and easily interpretable, but often less accurate than other classifiers.

A type of probabilistic classifier that uses Bayes' theorem to predict the class of an object based on its attributes. It assumes that all attributes are independent, which isn't always true.

The mathematical formula used in Naive Bayes to calculate the probability of an event based on prior knowledge and evidence.

The process of estimating the parameters of a Naive Bayes model using training data.

An unsupervised learning technique that aims to group data points into clusters based on their similarity. Similar items are grouped together, while dissimilar items are separated.

An unsupervised machine learning algorithm that aims to partition n observations into k clusters based on the distance of each observation to the cluster mean.

The initial point representing the mean value of a cluster in K-Means clustering. It's a starting point from which the algorithm iteratively adjusts the cluster centers.

This step in K-Means clustering involves calculating the distance between each data point and every cluster seed.

In this step, each data point is assigned to the cluster whose seed is the closest.

A process in K-Means clustering where the mean value of each cluster is recomputed after assigning data points to the closest clusters.

This criteria determines when the K-Means algorithm stops iterating. It can be based on no change in cluster assignments or reaching a predefined maximum number of iterations.

The commonly used squared Euclidean distance metric in K-Means clustering can be problematic when different dimensions have different scales. Rescaling the data can help, but it might not be suitable for nominal data.

K-Means clustering is not suitable for nominal data (data without inherent order) because the distance calculations that form the core of the algorithm are not meaningful for categorical data.

The overall within-cluster sum of squares (WCSS) is the total sum of squared distances between each data point and the centroid of its assigned cluster. In simpler terms, it measures the overall 'spread' of the data within the clusters.

In the context of clustering, the goal is to find a set of clusters that minimize the WCSS, which means finding clusters where the data points are tightly packed together around the centroid of each cluster.

An entity or agent that operates within an environment, making decisions and taking actions based on the current state, with the objective of receiving maximum rewards over time.

Variables that capture an agent's current position or situation in the environment. It defines the context for making decisions and taking actions

The actions an agent can take based on its current state. Each action affects the agent's interaction with the environment and has a consequence such as a reward or penalty.

Feedback or consequence provided to the agent based on its actions in a specific state. Rewards can be positive (encouraging) or negative (discouraging), guiding the agent towards desired behaviors.

A common algorithm in Reinforcement Learning that aims to find the optimal policy for an agent by learning the value of taking a specific action in a given state. It updates the agent’s knowledge about the environment and rewards, gradually improving its decision-making.