按方法、领域与证据探索科学。

General Circulation Model

A General Circulation Model (GCM), also called a Global Climate Model, is a three-dimensional numerical representation of the Earth's atmosphere, oceans, ice, and land surface that simulates physical processes governing weather and climate. Pioneered by Manabe and Wetherald in 1975, GCMs are the primary tools for under

machine learning

Generalized Additive Model

A generalized additive model, introduced by Trevor Hastie and Robert Tibshirani in 1986, extends the generalized linear model by replacing each linear term with a smooth, data-driven function of the predictor. This lets the model capture nonlinear relationships while preserving the additive, term-by-term interpretabili

deep learning

Generative Adversarial Network

A Generative Adversarial Network (GAN), introduced by Ian Goodfellow and colleagues in 2014, produces realistic synthetic data through the competition of two neural networks — a generator and a discriminator. It is widely used for image synthesis, data augmentation, and distribution estimation.

bioinformatics

Genome-wide association study

A genome-wide association study (GWAS) systematically tests hundreds of thousands to millions of single-nucleotide polymorphisms (SNPs) across the human genome for statistical association with a trait or disease. By comparing allele frequencies between cases and controls — or by regressing SNP genotypes on a quantitati

bioinformatics

Genome-wide association study in educational research

A genome-wide association study (GWAS) applied to educational research scans millions of single-nucleotide polymorphisms (SNPs) across the human genome to identify genetic variants statistically associated with educational outcomes such as years of schooling, degree attainment, or cognitive test scores. Large consortia

geoscience

Geochronological Dating

Geochronological dating is the determination of absolute ages of rocks and minerals using the decay of radioactive isotopes. Pioneered by Rutherford and Soddy (1902), this method provides numerical anchors for geological timescales and enables quantitative understanding of geological processes. Modern techniques (K-Ar,

geoscience

Geologic Mapping

Geologic mapping is the systematic observation and documentation of rock types, structures, and relationships exposed on the land surface. Pioneered by William Smith in 1799, this foundational field method remains essential for understanding subsurface geology, economic geology, hazard assessment, and paleoenvironmenta

geoscience

Geomechanical Modeling

Geomechanical modeling is the numerical simulation of stress and deformation in rock masses, integrating rock properties, pressure conditions, and geometric constraints. Rooted in classical mechanics (Coulomb, Mohr) but modernized by finite element and finite difference methods, this approach is essential for well inte

archaeology

Geometric Morphometrics

Geometric morphometrics is a quantitative analytical method that captures, analyzes, and compares the shapes of biological structures (bones, teeth, pottery) using coordinate data from landmarks and outlines. Developed by Fred Bookstein in the 1990s, GMM provides a rigorous statistical framework for studying shape vari

geoscience

Geophysical Inversion

Geophysical inversion is the process of using observed geophysical data to estimate subsurface properties and structures. Formalized by Tikhonov (1963) and expanded by Tarantola (1987), this mathematical framework solves the inverse problem: given measurements (gravity, magnetics, seismic, electrical), what subsurface

oceanography

Geostrophic Velocity

Geostrophic velocity is the current driven by balance between the pressure gradient force and the Coriolis force, derived from the thermal wind equation. In most of the ocean away from the equator and coastal boundaries, geostrophic balance is an excellent approximation to the actual flow. Developed by Harald Sverdrup

meteorology

Geostrophic Wind

Geostrophic wind balance is a fundamental concept in meteorology that describes the balance between the pressure gradient force and the Coriolis force in large-scale atmospheric flow. When this balance is achieved, wind blows parallel to isobars without acceleration—a condition observed in the free atmosphere away from

simulation

Global Sensitivity Analysis

Global sensitivity analysis (GSA) is a family of techniques that decompose the variance of a model's output across its input parameters, quantifying how much each input — and each combination of inputs — contributes to the total uncertainty in the result. Sobol's variance-based indices (2001), Morris's one-at-a-time (O

linguistics

Glottochronology

Glottochronology, or lexicostatistics, is a quantitative method in historical linguistics that estimates the time of divergence between related languages based on the proportion of shared cognates in their basic vocabularies. Developed by Morris Swadesh in 1950, the method assumes that core vocabulary items change at a

text mining

GloVe Embeddings

GloVe (Global Vectors for Word Representation) is a static word-embedding model introduced by Pennington, Socher and Manning (2014) that learns word vectors directly from global word-word co-occurrence statistics gathered across an entire corpus. The resulting vectors place semantically related words close together and

numerical methods

GMRES

GMRES (Generalized Minimal Residual) is an iterative method for solving large sparse non-symmetric or nonsymmetric linear systems Ax = b, developed by Saad and Schultz in 1986. It builds an orthonormal Krylov basis using Arnoldi's method and solves a least-squares problem to minimize residual at each iteration.

aerospace

GNSS RTK

Global Navigation Satellite System Real-Time Kinematic (GNSS RTK) is a high-precision positioning technique that uses carrier phase measurements from a reference receiver at a known location to correct the position estimates of a rover receiver in real time. Developed in the 1980s, RTK exploits spatial correlation of a

deep learning

GPT Fine-Tuning

GPT fine-tuning adapts pretrained autoregressive language models such as GPT-2/3/4 or LLaMA — introduced in OpenAI's 2019 work by Radford and colleagues — to domain-specific data or to instruction following via reinforcement learning from human feedback (RLHF) or DPO. It is used for instruction following, domain adapta

ophthalmology

GQL-15

The GQL-15 is a 15-item, disease-specific quality of life questionnaire designed to measure the impact of glaucoma on patients' daily functioning and psychological well-being. Developed by Nelson, Aspinall, and colleagues in the UK (1998), the GQL-15 emphasizes glaucoma-specific concerns—visual field loss, peripheral v

deep learning

Graph Attention Network

The Graph Attention Network (GAT), introduced by Veličković and colleagues in 2018, is a graph neural network variant that learns how much importance to assign to each neighbouring node through a self-attention mechanism. On heterogeneous neighbourhoods and relational classification it produces results superior to grap

deep learning

Graph Convolutional Network

Graph Convolutional Network (GCN) is a foundational deep learning architecture for graph-structured data, introduced by Thomas N. Kipf and Max Welling at ICLR 2017. It extends the convolution operation to irregular graph domains via a first-order spectral approximation, enabling each node to aggregate feature informati

network analysis

Graph Kernels

Graph kernels are positive semi-definite kernel functions that measure the similarity between two graphs by comparing their shared substructures — such as random walks, shortest paths, or subtree patterns. Introduced in a unified framework by Vishwanathan, Schraudolph, Kondor, and Borgwardt (2010), they bridge kernel m

deep learning

Graph Neural Network

A Graph Neural Network (GNN) is a deep learning method, popularised by Kipf and Welling in 2017 with the Graph Convolutional Network, that learns from the relationships in network (graph) structures made of nodes and edges. It is designed for data that is naturally relational, such as social networks, molecular structu

network analysis

Graph Neural Network (Network Analysis)

A Graph Neural Network (GNN) is a deep learning architecture that operates directly on graph-structured data by combining node features with structural information through iterative neighborhood message passing. The three canonical variants — the Graph Convolutional Network (GCN) introduced by Kipf and Welling in 2017,

deep learning

GraphRAG

GraphRAG is a retrieval-augmented generation approach that augments large language models with knowledge graphs to improve answer quality and factuality. Rather than retrieving flat text passages, GraphRAG constructs and queries structured knowledge graphs extracted from documents, providing rich contextual information

astronomy

Gravitational Microlensing

Gravitational microlensing is an observational technique that exploits Einstein's prediction that massive objects bend light. When a star or planet passes in front of a distant star from our perspective, its gravity acts as a lens, magnifying and distorting the background star's light. First proposed by Bohdan Paczynsk

architecture

Green Building Rating System

Green Building Rating Systems are standardized frameworks for assessing and certifying the environmental performance and sustainability of buildings. The most widely known is LEED (Leadership in Energy and Environmental Design), established by the U.S. Green Building Council in 1998. Similar systems exist globally (BRE

manufacturing

Griffith Fracture Mechanics

Griffith's theory of brittle fracture explains how small flaws or cracks in materials grow unstably, leading to sudden catastrophic failure. Formulated by Alan A. Griffith in 1921 through experiments on glass fibers, this theory balances the elastic energy released by crack growth against the surface energy required to

quantum computing

Grover's Algorithm

Grover's Algorithm is a quantum algorithm for searching an unsorted database, offering a quadratic speedup over classical linear search. Proposed by Lov Grover in 1996, it exploits quantum superposition and amplitude amplification to find a target item among N items in O(√N) queries, compared to the classical O(N) requ

deep learning

GRU

The Gated Recurrent Unit (GRU) is a gated recurrent neural network cell introduced by Cho and colleagues in 2014 that captures long-range dependencies in sequential data using update and reset gates, achieving performance comparable to LSTM with fewer parameters.

control theory

H-infinity Control

H-infinity (H∞) control is a robust control method that minimizes the worst-case gain from disturbances to controlled outputs, formulated as a minimax optimization problem. Pioneered by Zames in the early 1980s, H∞ control provides a principled way to design feedback controllers that tolerate model uncertainty, unmodel

text mining

Hallucination Detection

Hallucination detection is a natural-language-processing pipeline that measures whether the output of a language model is consistent with a reference source document or with verifiable facts. Formalised as a faithfulness evaluation task by Maynez et al. (2020) and extended to a zero-resource black-box setting by Manaku

astronomy

Halo Occupation Distribution

Halo Occupation Distribution (HOD) modeling is a framework for relating observed galaxy clustering to the distribution of galaxies within dark matter halos. Developed by Jia Peacock and others around 2000, HOD provides a flexible, physically motivated approach to interpreting galaxy surveys and understanding how galaxi

numerical methods

Halstead Complexity

Halstead Complexity Metrics are a set of static code analysis measures developed by Maurice Halstead in 1977 that quantify software quality using operator and operand counts. Metrics like program volume, difficulty, and effort estimate code complexity, maintainability, and defect likelihood from source code structure a

model evaluation

Hamming Loss

Hamming loss measures the fraction of labels that are incorrectly predicted in multi-label classification. It counts the number of label mistakes divided by the total number of labels, providing a simple metric for multi-label problems.

civil engineering

Hardy Cross Method

The Hardy Cross method is an iterative technique for solving steady-state flow distribution in pipe networks, originally developed for water distribution systems. Introduced by Hardy Cross in 1936, this method balances flow continuity and pressure head constraints through successive iterations, making it ideal for hand

oceanography

Harmful Algal Bloom Monitoring

Harmful algal bloom (HAB) monitoring is an integrated approach combining satellite remote sensing, in situ observations, and predictive modeling to detect, track, and forecast toxic algal outbreaks in marine and freshwater systems. HAB monitoring has become essential for public health protection, as certain algal speci

music information retrieval

Harmonic Analysis in Music

Harmonic analysis is the computational study of chord progressions, harmonic function, and tonal relationships in music. Formalized for audio by Pardo and Birmingham (2002), it goes beyond simple chord identification to interpret harmonic role and structure. Harmonic analysis is essential for music theory education, co

electrical engineering

Harmonic Distortion Analysis

Harmonic distortion analysis quantifies the deviation of voltage or current waveforms from sinusoidal shape due to nonlinear loads. Using Fourier decomposition, engineers separate the waveform into its fundamental frequency and harmonic components (integer multiples of 50 or 60 Hz). Harmonic analysis is critical for as

computer vision

Harris Corner Detection

The Harris corner detector, introduced by Chris Harris and Mike Stephens in 1988, is a foundational method for identifying corners and interest points in digital images. Harris corners are points where two edges meet at a significant angle, making them stable and repeatable features for image analysis, matching, and 3D

quantum computing

Hartree-Fock Method

The Hartree-Fock (HF) method is a foundational self-consistent field approach for solving the many-electron Schrödinger equation. Developed independently by Douglas Hartree and Vladimir Fock in the late 1920s, it approximates the ground state by assuming electrons move in an average field generated by all other electro

text mining

Hate Speech Detection

Hate speech detection is a natural-language-processing task that automatically identifies hateful, offensive, or harmful text on social media and online platforms. The task was sharpened by Davidson and colleagues (2017), who showed why separating genuine hate speech from merely offensive language is a hard, distinct c

machine learning

HDBSCAN

HDBSCAN (Hierarchical Density-Based Spatial Clustering of Applications with Noise) is a density-based clustering algorithm introduced by Campello, Moulavi, and Sander in 2013. It extends DBSCAN by building a full hierarchy of density-based clusters across all density scales and then extracting a stable flat partition,

applied physics

Head-Related Transfer Function

The Head-Related Transfer Function (HRTF) describes how the human head, ears, and torso filter sound from different directions. HRTFs capture the acoustical changes that occur as sound travels around the head to reach each ear, enabling the perception of sound location in 3D space. Measured or modeled HRTFs are essenti

biomechanics

Heart Rate Variability

Heart rate variability (HRV) analysis quantifies the variation in time intervals between consecutive heartbeats as a window into autonomic nervous system function and cardiovascular health. Formalized by the European Society of Cardiology Task Force in 1996, HRV metrics are now standard in cardiology, physiology, and s

geophysics

HEC-RAS

HEC-RAS (Hydrologic Engineering Center River Analysis System) is a hydraulic modeling software developed by the US Army Corps of Engineers that computes water surface elevation and velocity in open channels and floodplains, and depicts inundation extent and depth. Since its introduction in 1995, HEC-RAS has become the

particle physics

HEP Track Reconstruction

Track reconstruction is the process of identifying and measuring the trajectories of charged particles through a detector, providing momentum and impact parameter information essential for particle identification, vertex reconstruction, and physics analysis in high-energy physics experiments.

genetics

Hi-C Analysis

Hi-C (High-Chromosome Conformation Capture) is a technique and associated computational methods for mapping the 3D architecture of the genome within cells. Developed by Lieberman-Aiden and Dekker in 2009, Hi-C identifies physical interactions between genomic regions that may be distant in linear sequence but spatially

machine learning

Hierarchical Clustering

Hierarchical clustering is an unsupervised method that groups observations into nested clusters and draws the result as a dendrogram, so the number of clusters need not be fixed in advance. Its agglomerative form rests on the objective-function grouping criterion introduced by Joe Ward in 1963.

reliability engineering

Highly Accelerated Life Testing

Highly Accelerated Life Testing (HALT) is a methodology for rapidly identifying design weaknesses and determining the margin between normal operating conditions and product failure. By applying extreme but non-destructive stress profiles (thermal, vibration, etc.), HALT accelerates the failure clock to reveal latent de

signal processing

Hilbert-Huang Transform

The Hilbert-Huang Transform (HHT) is an adaptive, data-driven method for analyzing non-linear and non-stationary time series, introduced by Norden E. Huang and colleagues in 1998. It combines Empirical Mode Decomposition (EMD), which decomposes a signal into intrinsic mode functions (IMFs), with the Hilbert spectral an

computer vision

Histogram Equalization

Histogram equalization is an image preprocessing technique that redistributes pixel intensities to improve contrast and visibility of details. By spreading the histogram of pixel values evenly across the available range, histogram equalization enhances images with poor contrast, making features more visually distinct a

genetics

HKA Test

The Hudson-Kreitman-Aguade (HKA) test is a statistical method that tests for neutral evolution by comparing levels of within-population polymorphism and between-population divergence at multiple loci. Developed by Hudson, Kreitman, and Aguade in 1987, this test uses the principle that neutral loci should show expected

cryptography

HMAC

HMAC (Hash-Based Message Authentication Code) is a cryptographic algorithm for authenticating messages using a secret key and a hash function. Standardized in RFC 2104 (1997), HMAC can be combined with any cryptographic hash function (SHA-256, SHA-3, etc.) to create a message authentication code (MAC). HMAC provides bo

bioinformatics

HMMER Profile Search

HMMER profile search identifies distant protein sequence homologs using probabilistic models of protein families, known as profile Hidden Markov Models (HMMs). Developed by Eddy and colleagues, this method captures sequence variation patterns within protein families and detects homologs with far greater sensitivity tha

biomechanics

Hodgkin-Huxley Model

The Hodgkin-Huxley model is a mathematical description of how action potentials in neurons are generated by the flow of sodium and potassium ions across the cell membrane. Developed by Alan Hodgkin and Andrew Huxley in 1952, it is a foundational model in neuroscience and earned them the Nobel Prize, establishing quanti

mining engineering

Hoek-Brown Criterion

The Hoek-Brown Criterion, developed by Evert Hoek and E. T. Brown starting in 1980, is an empirical failure criterion that predicts the shear strength of rock masses as a function of confining pressure. It accounts for rock quality (via the Geological Strength Index, GSI) and thus bridges laboratory rock mechanics and

applied physics

Hohmann Transfer

The Hohmann transfer is a maneuver that transfers a spacecraft between two circular orbits using two impulsive burns (velocity changes). Introduced by German engineer Walter Hohmann in 1925, it is the most fuel-efficient method for coplanar orbital transfers when the transfer time is not severely constrained. The trans

aerospace

Holtrop-Mennen Method

The Holtrop-Mennen Method is an empirical regression-based technique for predicting total ship resistance from geometric parameters and operating conditions. Developed by Jelte Holtrop and Gert Mennen in 1982, the method decomposes total resistance into friction, pressure, wave-making, and form drag components, each es

bioinformatics

Homology Modeling

Homology modeling, also called comparative modeling, predicts the three-dimensional structure of a protein using an experimentally-solved structure of a homologous protein as a template. Introduced by Sali and Blundell in 1993, this method exploits the principle that homologous proteins share similar spatial structures