Data Science Glossary | Practice Test / Quiz / MCQs

Home > Data Science > Quizzes > Data Science Glossary

Data Science Glossary

Fast practice, instant feedback. Timer auto-submits when time’s up.

Avg score: 1% • Most missed: “Greedy Algorithm”

Study First Challenge a Friend Stats

Data Science Glossary

Time left 00:00

25 Questions

1. Big Data

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.A class of machine learning algorithms that are inspired by the neuron architecture of the human brain. Typically - a (multi-layer) ANN consists of a layer of input nodes - a layer of output nodes - and hidden layers in-between. The nodes are connected by weighted links that can be interpreted as the neuron-connections by axons of different strengths. The simplest version of an ANN is a single-layer perceptron.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

2. Correlation analysis - Matthews Correlation Coefficient (MCC)

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.A class of machine learning algorithms that are inspired by the neuron architecture of the human brain. Typically - a (multi-layer) ANN consists of a layer of input nodes - a layer of output nodes - and hidden layers in-between. The nodes are connected by weighted links that can be interpreted as the neuron-connections by axons of different strengths. The simplest version of an ANN is a single-layer perceptron.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.

3. Objective function

A class of machine learning algorithms that are inspired by the neuron architecture of the human brain. Typically - a (multi-layer) ANN consists of a layer of input nodes - a layer of output nodes - and hidden layers in-between. The nodes are connected by weighted links that can be interpreted as the neuron-connections by axons of different strengths. The simplest version of an ANN is a single-layer perceptron.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

4. Artificial Neural Networks (ANN)

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.A class of machine learning algorithms that are inspired by the neuron architecture of the human brain. Typically - a (multi-layer) ANN consists of a layer of input nodes - a layer of output nodes - and hidden layers in-between. The nodes are connected by weighted links that can be interpreted as the neuron-connections by axons of different strengths. The simplest version of an ANN is a single-layer perceptron.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

5. Monte Carlo simulation

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An iterative sampling method for solving deterministic models. Random numbers or variables from a particular probability distribution are used as input variables for uncertain parameters to compute the response variables.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

6. K-nearest neighbors algorithms

Find the k-points that are closest to a point of interest based on their attributes using a certain distance measure (e.g. - Euclidean distance). K-nearest neighbors algorithms are being used in many different contexts: Non-parametric density estimation - missing value imputation - dimensionality reduction - and classifiers in supervised and unsupervised pattern classification and regression problems.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.

7. Active Learning

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.A variant of the on-line learning machine learning architecture where feedback about the ground truth class labels of unseen data can be requested if the classification is uncertain. New training data that was labeled can then be used to update the model as in on-line learning.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

8. k-fold Cross-validation

The data is split into k subsets - then a prediction/classification model is trained k times - each time holding one subset as test set - training the model parameters using the remaining k-1. Finally - cross-validation error is evaluated as the average error out of all k training models.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

9. Support Vector Machine (SMV)

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.A classification method that tries to find the hyperplane which separates classes with highest margin. The margin is defined as the minimum distance from sample points to the hyperplane. The sample point(s) that form margin are called support vectors and eventually establish the SVM model.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

10. Jittering

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.A sampling technique that can be used to measure the stability of a given statistical model (classifiction/regression/clustering). In jittering - some noise is added to sample data points - and then a new model is drawn and compared to the original model. (submitted by Vahid Mirjalili) An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

11. Correlation analysis

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.Describes and quantifies the relationship between two independent variables. Typically - in case of a positive correlation both variables have a tendency to increase - and in the case of negative correlation - one variable increases while the other variable increases. It is important to mention the famous quotation "correlation does not imply causation".There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.

12. Resubstitution error

Represents the classification error rate on the training dataset (the dataset that was used to train the classifier). The performance of a classifier cannot be directly deduced from resubstitution error alone - but it becomes a useful measure for calculating the generalization error.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

13. DBSCAN

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.A variant of a density-based clustering algorithm that identifies core points as regions of high-densities based on their number of neighbors (> MinPts) in a specified radius (ε). Points that are below MinPts but within ε are specified as border points; the remaining points are classified as noise points.

14. ####iid

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.The abbreviation "iid" stands for "independent and identically distributed" and describes random variables that are independent from one another and are drawn from a similar probability distribution. Independence means that the probability of one observation does not affect the probability of another variable (e.g. - time series and network graphs are not independent). One popular example of iid would be the tossing of a coin: One coin toss does not affect the outcome of another coin toss - and the probability of the coin landing on either "heads" or "tails" is the same for every coin toss.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

15. Central Limit Theorem

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.A theorem in the field of probability theory that expresses the idea that the distribution of sample means (from independent random variables) converges to a normal distribution when the sample size approaches infinity. There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

16. Local Outlier Factor (LOF)

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.A density-based anomaly detection technique for outlier identification. The LOF for a point p refers to the average "reachability distance" towards its nearest neighbors. Eventually - the points with the largest LOF values (given a particular threshold) are identified as outliers.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

17. Cost function

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.(synonymous to loss function) - is a special case of an objective function - i.e. - a function that is used for solving optimization problems. A cost function can take one or more input variables and the output variable is to be minimized. A typical use case for cost functions is parameter optimization.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

18. Least Squares fit - linear

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.A linear regression technique that fits a straight line to a data set (or overdetermined system) by minimizing the sum of the squared residuals - which can be the minimized vertical or perpendicular offsets from the fitted line.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.

19. Rule-based classifier

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.Classifiers that are based on one or more "IF ... THEN ..." rules. Rule-based classifiers are related to decision trees and can be extracted from the latter. If the requirements for a rule-based classifier (mutually exclusive: at most one rule per sample; mutuallyy exhaustive: at least one rule per sample) are violated - possible remedies include the addition of rules or the ordering of rules.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

20. Fuzzy C-Means Clustering

A soft clustering algorithm in which each sample point has a membership degree to each cluster; in hard (crisp) clustering - membership of each point to each cluster is either 0 or 1. Fuzzy C-Means considers a weight matrix for cluster memberships - and minimizes sum squared error (SSE) of weighted distances of sample points to the cluster centroids.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

21. Independent Component Analysis ICA)

Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.A statistical signal-processing technique that decomposes a multivariate dataset of mixed - non-gaussian distributed source signals into independent components A popular example is the separation of overlapping voice samples -- the so-called "cocktail party problem".

22. Expectation Maximization algorithm - EM

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.Atechnique to estimate parameters of a distribution based on the Maximum Likelihood Estimate (MLE) that is often used for the imputation of missing values in incomplete datasets. After the EM algorithm is initialized with a starting value - alternating iterations between expectation and maximization steps are repeated until convergence. In the expectation step - parameters are estimated based on the current model to impute missing values. In the maximization step - the log-likelihood function of the statistical model is to be maximized by re-estimating the parameters based on the imputed values from the expectation step.

23. Hyperparameters

The parameters of a classifier or estimator that are not directly learned in the machine learning step from the training data but are optimized separately (e.g. - via Grid Search). The goal of hyperparameter optimization is to achieve good generalization of a learning algorithm and to avoid overfitting to the training data.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

24. Perceptron

There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.A simple Artificial Neural Network algorithm that consists of only two types of nodes: Input nodes and output nodes connected by weighted links. Perceptrons are being used as linear classifiers in supervised machine learning tasks.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.

25. Machine learning

An assessment metric for clustering or binary classification analyses that represents the correlation between the observed (ground truth) and predicted labels. MCC can be directly computed from the confusion matrix and returns a value between -1 and 1.A set of algorithmic instructions for discovering and learning patterns from data e.g. - to train a classifier for a pattern classification task.Mathematical function that are used for problem-solving and optimization tasks. Depending on the task - the objective function can be omtpimized through minimization (cost or loss functions) or maximization (reward function). A typical application of an objective function in pattern classification tasks is to minimize the error rate of a classifier.There are many different - controversial interpretations and definitions - one refers to data as "Big Data" if its volume and complexity are of a magnitude that the data cannot be processed by "conventional" computing platforms anymore; storage space - processing power - and database structures are typically among the limiting factors.

⚡ Recently practiced quizzes in this topic

Live quiz activity

Data Analytics Practice Test Big Data & Analytics NASSCOM Certification Practice Test PySpark Practice Test Questions Basic Data Analytics and Visualization Practice Test (Tableau) Data Analysis with Python Data Science Exam #1 Data Analytics and Visualization Practice Test Pega Certified System Architect (PCSA) Study Guide Data Science Basics / Data Scientist Toolbox Pega Data Scientist

25 Questions

❤ If you liked Fatskills, consider supporting us by checking out The Life Manuals You Never Got.

About | Explore | User Guide | Topics | Subjects | Doubt Solver | Career Aptitude Test | Answers | Free Tools | OSHA Basics Quiz | What Should We Know?
Privacy | Terms |

Without work one finishes nothing. - Ralph Waldo Emerson
© 2026 Fatskills.com

All trademarks, logos and brand names are the property of their respective owners. All company, product and service names used in this website are for identification purposes only. Use of these names, trademarks and brands does not imply endorsement.

25 Questions

❤ If you liked Fatskills, consider supporting us by checking out The Life Manuals You Never Got.

About | Explore | User Guide | Topics | Subjects | Doubt Solver | Career Aptitude Test | Answers | Free Tools | OSHA Basics Quiz | What Should We Know? Privacy | Terms |

Without work one finishes nothing. - Ralph Waldo Emerson© 2026 Fatskills.com

All trademarks, logos and brand names are the property of their respective owners. All company, product and service names used in this website are for identification purposes only. Use of these names, trademarks and brands does not imply endorsement.

About | Explore | User Guide | Topics | Subjects | Doubt Solver | Career Aptitude Test | Answers | Free Tools | OSHA Basics Quiz | What Should We Know?
Privacy | Terms |

Without work one finishes nothing. - Ralph Waldo Emerson
© 2026 Fatskills.com