Skip to content

Concept Glossary: D

This page lists English glossary entries for this letter. Entry bodies are assembled from term source files.

data

  • Meaning: Data is information used in different roles: training examples for learning, evaluation material for checking, retrieved evidence for services, user input, state, or operational records.
  • Why it matters: The same word data can mean different things in model training and deployed services. Readers need to distinguish what the information is being used for before judging quality, responsibility, or system behavior.
  • Related concepts: model input, model output, supervised learning label, source data, provenance
  • Core Section: P1-4.2
  • Appears in: P1-2.3, P4-1.1, P1-3.2, P1-14.1

data leakage

  • Meaning: Data leakage occurs when information unavailable at prediction time enters training or evaluation and makes performance look unfairly good.
  • Why it matters: Leakage breaks the boundary of a fair experiment. A high score may reflect hidden future information or target-like hints rather than real generalization, so data splitting and preprocessing order must be inspected.
  • Related concepts: dataset, model validation, target
  • Core Section: P2-12.3
  • Appears in: P3-4.2, P3-9.7, P3-9.13, P4-7.1, P4-7.2, P4-7.3

data modeling

  • Meaning: Data modeling is the process of turning raw records into samples, features, comparisons, and output structures that can answer a specific question.
  • Why it matters: Many records do not automatically form a learnable problem. Data modeling fixes what counts as one case, what should be compared, and what output form closes the task.
  • Related concepts: dataset, sample unit, comparison report, output structure, task definition
  • Core Section: P3-1.1
  • Appears in: P3-index, P3-1.2, P3-1.3, P3-2.1, P3-2.2, P3-3.1, P3-3.2, P3-summary

data science

  • Meaning: Data science is a broad workflow for collecting, cleaning, representing, modeling, and interpreting data so it can answer questions and support decisions.
  • Why it matters: Data modeling is only one part of this larger workflow. Separating the broader field from the narrower modeling step keeps collection, cleaning, exploration, modeling, interpretation, and decision support from being treated as the same task.
  • Related concepts: data modeling, dataset, preprocessing, exploration, decision
  • Core Section: P3-1.1
  • Appears in: P3-index, P3-summary, P4-1.2

data structure

  • Meaning: A data structure organizes data so particular operations, such as lookup, insertion, traversal, or relationship following, become natural.
  • Why it matters: The right structure depends on which questions are asked often. Lists, dictionaries, trees, and graphs differ because they support different access patterns and costs.
  • Related concepts: dataset, feature, vectorization
  • Core Section: P2-9.1
  • Appears in: P2-9.2, P2-9.3, P2-9.4

dataset

  • Meaning: A dataset is a working collection of samples and variables prepared for training or evaluation. It includes choices about inputs, targets, sample boundaries, and splits.
  • Why it matters: The same raw table can become different learning problems depending on how the dataset is built. Dataset design often shapes performance and interpretation before model choice does.
  • Related concepts: data structure, feature, target, sample unit, task definition
  • Core Section: P2-12.3
  • Appears in: P2-12.1, P2-12.2, P3-2.1, P3-2.2, P3-3.2, P3-4.2, P3-9.2, P3-summary, P4-5.2

DBSCAN

  • Meaning: DBSCAN is a density-based clustering method that treats sufficiently dense regions of points as clusters and can leave low-density points as noise or outliers.
  • Why it matters: DBSCAN gives a different intuition from k-means. Instead of choosing the number of clusters first and assuming groups around centroids, it asks whether points are densely connected under the current distance rule.
  • Related concepts: clustering, cluster, density, outlier
  • Core Section: P4-17.1
  • Appears in: P4-17.3

decision boundary

  • Meaning: A decision boundary is the line, surface, or higher-dimensional criterion that separates class regions in the input space. In logistic regression, it is usually read as the place where the linear score \(z\) equals the classification criterion.
  • Why it matters: The boundary helps explain why an input moved to one class rather than another and why near-boundary cases often become review targets. It also keeps model score, threshold choice, and applied class region separate when an operating threshold changes.
  • Related concepts: logistic regression, threshold, classification, hyperplane
  • Core Section: P4-11.2
  • Appears in: P4-11.3, P5-1.2

decision tree

  • Meaning: A decision tree is a tree-based model that reaches predictions by asking a sequence of condition questions and following branches to a leaf.
  • Why it matters: Decision trees are intuitive for classification and regression, but deep trees can memorize accidental training patterns. They show the tradeoff between interpretability and overfitting risk.
  • Related concepts: threshold, overfitting, classification, regression
  • Core Section: P4-14.1
  • Appears in: P4-index, P4-3.2, P4-4.2, P4-5.1, P4-14.1, P4-14.2

decoder

  • Meaning: A decoder is the component that generates the next token or output sequence from current context and internal representations.
  • Why it matters: Decoders explain output generation in encoder-decoder and GPT-style models. They separate the role of reading input from the role of continuing or producing output step by step.
  • Related concepts: encoder, Encoder-Decoder, GPT
  • Core Section: P1-11.3
  • Appears in: P1-11.2, P6-5.1, P6-6.1

deep learning

  • Meaning: Deep learning is a machine-learning approach that uses multi-layer neural networks to learn intermediate representations and prediction rules together.
  • Why it matters: Deep learning marks the move from hand-designed features toward learned representations. It is a specific neural-network-centered approach inside machine learning, not a synonym for all AI.
  • Related concepts: representation learning, parameter, feature
  • Core Section: P1-2.3
  • Appears in: P1-1.3, P4-1.1, P4-index

derivative

  • Meaning: A derivative is the instantaneous rate of change: how much output changes when input changes by a very small amount.
  • Why it matters: Derivatives connect function behavior to optimization and learning. They prepare readers to understand gradients and why training follows directions that reduce loss.
  • Related concepts: rate of change, gradient, partial derivative
  • Core Section: P2-4.3
  • Appears in: P2-4.4, P2-6.1, P2-6.3

diffusion model

  • Meaning: A diffusion model is a generative model family that starts from a noisy state and gradually denoises it into a meaningful image or representation.
  • Why it matters: Diffusion models show that generation is not always next-token continuation. Image generation often depends on iterative refinement over many steps.
  • Related concepts: generation, sampling, probabilistic model
  • Core Section: P1-10.2

dimension

  • Meaning: A dimension is the number of values or coordinate axes used to represent a vector or data point.
  • Why it matters: Dimensions help readers understand vector shape, matrix shape, feature count, and why high-dimensional search or visualization becomes difficult.
  • Related concepts: vector, matrix, tensor
  • Core Section: P2-3.2
  • Appears in: P2-3.3, P2-11.1, P2-11.2, P4-18.1

dimensionality reduction

  • Meaning: Dimensionality reduction moves high-dimensional data into fewer axes while trying to preserve important structure.
  • Why it matters: It makes complex data easier to visualize or summarize, but the reduced view is not a perfect copy of all original information.
  • Related concepts: unsupervised learning, dimension, representation
  • Core Section: P4-18.1
  • Appears in: P1-8.2, P4-index, P4-2.2, P4-3.2, P4-7.2, P4-7.3, P4-7.4, P4-18.2

distance

  • Meaning: Distance is a numerical measure of how far apart two vectors are in a representation space.
  • Why it matters: Distance turns “close” or “similar” into a computable comparison rule for ranking candidates in search and retrieval systems.
  • Related concepts: similarity, nearest neighbor, vector space, similarity search
  • Core Section: P1-13.2
  • Appears in: P1-13.2, P2-3.2, P4-6.4, P4-7.2, P4-7.3, P4-12.2, P4-12.3, P6-3.2

density

  • Meaning: Density describes how tightly data points are packed in a region of a representation space.
  • Why it matters: Not every cluster is round or evenly sized. Density helps explain grouping methods that find crowded regions and leave sparse points aside.
  • Related concepts: clustering, distance, similarity
  • Core Section: P4-17.1
  • Appears In:

distributed representation

  • Meaning: A distributed representation expresses a word or object as a vector spread across many numerical dimensions instead of one symbol.
  • Why it matters: Distributed representations support flexible generalization because related items can occupy nearby positions through many partially shared dimensions.
  • Related concepts: embedding, vector, corpus
  • Core Section: P1-11.1
  • Appears in: P1-9.3, P1-13.1, P5-2.2, P6-3.1, P6-20.2

distribution

  • Meaning: A distribution describes where values concentrate and how they spread across a dataset.
  • Why it matters: Distributions reveal skew, outliers, multimodal structure, and variation that a single average can hide.
  • Related concepts: probability, mean, variance
  • Core Section: P2-5.2
  • Appears in: P2-5.3, P2-13.1, P2-13.2, P4-6.3

data distribution

  • Meaning: The pattern describing what kinds of values appear in data and how frequently or widely they appear.
  • Why it matters: It matters because model behavior depends not only on individual examples but also on the distribution from which those examples come.
  • Related concepts: Read together with the surrounding glossary entries and the linked Sections for this term.
  • Core Section: P2-5.2
  • Appears in: P2-5.5

determinism

  • Meaning: The property that the same conditions and inputs lead to the same result.
  • Why it matters: It helps separate reproducible computation from stochastic or nondeterministic behavior in training, generation, and software execution.
  • Related concepts: Read together with the surrounding glossary entries and the linked Sections for this term.
  • Core Section: P1-3.1
  • Appears in: P1-6.2

dense vector

  • Meaning: A vector in which many dimensions carry nonzero values, often used to represent meaning in embeddings.
  • Why it matters: It distinguishes learned semantic representations from sparse identifier-like encodings such as one-hot vectors.
  • Related concepts: Read together with the surrounding glossary entries and the linked Sections for this term.
  • Core Section: P1-11.1

deep reinforcement learning

  • Meaning: A reinforcement learning approach that uses deep neural networks to represent policies, value functions, or environment signals.
  • Why it matters: It connects reinforcement learning problems to deep learning representation power, especially when states or observations are too complex for simple tables.
  • Related concepts: Read together with the surrounding glossary entries and the linked Sections for this term.
  • Core Section: P1-8.3

delayed reward

  • Meaning: A reward signal that appears only after several steps rather than immediately after the action that helped cause it.
  • Why it matters: It explains why reinforcement learning must consider long-term consequences instead of choosing only what looks best right now.
  • Related concepts: Read together with the surrounding glossary entries and the linked Sections for this term.
  • Core Section: P1-8.3