按方法、领域与证据探索科学。

Space Syntax Analysis

Space Syntax Analysis is a quantitative method for assessing spatial configuration in buildings and urban environments through graph-based representations. Developed by Bill Hillier and Julienne Hanson in the 1980s, it quantifies how spatial layout affects human movement, visibility, and social interaction.

deep learning

Spatial-Temporal GCN

Spatial-Temporal Graph Convolutional Networks (ST-GCN) is an architecture introduced by Yan et al. in 2018 for skeleton-based action recognition. By modeling human skeletons as graphs where joints are nodes and bones are edges, ST-GCN applies graph convolutions across space and time to recognize actions from skeleton s

aerospace

Specific Excess Power

Specific excess power (Ps) is a metric that quantifies the rate of change of energy per unit weight, representing how quickly an aircraft can trade speed for altitude (or vice versa) at a given flight condition. Developed by John Boyd in the 1970s as part of energy maneuverability theory, Ps is essential for assessing

model evaluation

Specificity

Specificity measures the proportion of actual negative cases that were correctly identified as negative by the classifier. It answers the question: 'Of all the cases that were truly negative, how many did we correctly reject?' Specificity is complementary to recall and is essential when false positives are costly.

meteorology

Spectral Bin Microphysics

Spectral bin microphysics is a detailed cloud microphysical modeling approach that explicitly represents the particle size distribution (PSD) by dividing particles into discrete size bins. Rather than assuming a fixed shape for the PSD, bin models track the number and mass of particles in each size category, allowing d

machine learning

Spectral Clustering

Spectral Clustering is a graph-based unsupervised learning algorithm, formalized by Ng, Jordan, and Weiss in 2002, that maps data points into a low-dimensional eigenspace derived from the similarity graph's Laplacian before applying k-means. This spectral embedding makes it possible to recover clusters of arbitrary sha

numerical methods

Spectral Methods

Spectral Methods are high-order numerical techniques for solving differential equations using global polynomial expansions (e.g., Fourier or Legendre series) rather than local piecewise polynomials. Developed by Steven Orszag in the 1960s for turbulence simulation, they offer exponential convergence for smooth problems

text mining

Speculation Detection

Speculation detection, also known as hedging analysis, is a natural-language-processing task that identifies epistemic uncertainty markers — words and phrases such as 'may', 'possibly', 'it is suggested that' — within scientific, biomedical, and news texts. Formalised by Hyland (1996) for scientific writing and benchma

linguistics

Speech Act Theory

Speech Act Theory is a framework in pragmatics developed by J. L. Austin and refined by John Searle, analyzing language as action. The core insight is that utterances are not merely vehicles for propositions but acts with pragmatic effects: 'I pronounce you married' creates a marriage; 'Please close the door' issues a

acoustics

Speech Intelligibility

Speech intelligibility is a quantitative measure of how well listeners understand spoken content in acoustic environments. Formalized by Steeneken and Houtgast in 1980 with the Speech Transmission Index (STI), intelligibility metrics combine room acoustic parameters (RT60, noise, clarity) to predict listener comprehens

text mining

Spelling and Grammar Check

Spelling and grammar checking is a text-mining task that detects spelling mistakes and grammatical errors in text and proposes corrections. Building on Naber's rule-based style and grammar checker (2003) and Norvig's statistical spelling corrector (2009), it is used for data-quality assessment and text normalisation be

ensemble learning

Stacked Generalization

Stacked generalization, or stacking, is a two-level ensemble method where base-level classifiers are trained on the original data, and a meta-learner is trained on the predictions of the base classifiers. The meta-learner learns how to best combine base predictions rather than using fixed aggregation rules. Introduced

machine learning

Stacking

Stacking, or stacked generalization, is an ensemble method introduced by David Wolpert in 1992 that combines the outputs of several different base models (Level-0) through a separate meta-model (Level-1). Unlike bagging and boosting, it deliberately uses heterogeneous model types, and it is the standard final-stage str

text mining

Stance Detection

Stance detection is a natural-language-processing task that decides the position a text takes toward a specific claim, event, or topic — labelling it as favor, against, or neutral. Formalised by Mohammad et al. in the SemEval-2016 Task 6 shared task, it differs from plain sentiment analysis because the label is always

geophysics

Standardized Precipitation Evapotranspiration Index

The Standardized Precipitation Evapotranspiration Index (SPEI) is a climate index that combines precipitation and temperature (via reference evapotranspiration) to characterize water deficits and droughts. Developed by Vicente-Serrano and colleagues in 2010, SPEI extends the SPI framework to account for the combined ef

geophysics

Standardized Precipitation Index

The Standardized Precipitation Index (SPI) is a climate index that quantifies precipitation anomalies relative to historical norms, standardized to account for differences in precipitation climatology across regions. Introduced by McKee, Doesken, and Kleist in 1993, SPI has become a primary tool for drought detection a

thermodynamics

State of Charge

State of Charge (SOC) is the amount of energy available in a battery or energy storage system, expressed as a percentage of its maximum capacity. Accurate SOC estimation is critical for safe operation: underestimating SOC can cause unsafe discharges, overestimating can cause overcharging. SOC estimation combines curren

thermodynamics

State of Health

State of Health (SOH) quantifies battery degradation by measuring how much capacity and power capability have been lost due to aging. SOH is expressed as a percentage (100% = new, 80% = end of life for many applications). Tracking SOH enables predictive maintenance, end-of-life detection, and accurate range/power predi

cryptography

Static Application Security Testing

Static Application Security Testing (SAST) is a security analysis technique that examines source code or compiled binaries without executing the program to identify vulnerabilities, code quality issues, and security flaws. Developed in the 2000s, SAST analyzes code structure, data flow, and control flow to detect poten

software engineering

Static Code Analysis

Static code analysis automatically examines source code without execution, detecting potential bugs, security vulnerabilities, code smells, and style violations. Pioneered by Engler and Pugh (2001), automated analysis tools scan codebases at scale, identifying defect patterns faster than manual review. Organizations in

electrical engineering

Static Timing Analysis

Static Timing Analysis (STA) is a non-simulation method for verifying that digital circuits meet timing constraints (clock frequencies, setup/hold times, propagation delays). Introduced systematically by Bhatnagar et al. in the 1990s, STA computes worst-case and best-case path delays by analyzing logic paths without si

thermodynamics

Stefan-Maxwell Diffusion

The Stefan-Maxwell diffusion equation describes how multiple chemical species diffuse through each other in a mixture, accounting for interactions between all species pairs. Unlike Fick's law, which assumes species diffuse independently, Stefan-Maxwell theory captures the coupling that occurs when species with differen

astronomy

Stellar Population Synthesis

Stellar population synthesis is a technique for modeling the integrated light from a galaxy by summing the contributions of all individual stars formed at different times and with different masses and metallicities. Developed systematically by Bruzual and Charlot (2003), this approach enables estimation of fundamental

computer vision

Stereo Matching

Stereo matching is a computer vision technique for recovering depth information by finding corresponding points between a pair of stereo images (taken from slightly different viewpoints). By locating the same scene feature in both images and measuring the disparity (horizontal shift), stereo matching reconstructs 3D st

mining engineering

Stereographic Slope Analysis

Stereographic projection is a graphical method for analyzing slope stability by representing the three-dimensional orientation of discontinuities (joints, bedding, faults) and the pit slope on a two-dimensional stereographic net (stereonet). The method enables rapid visual identification of potentially unstable slope g

network analysis

Stochastic Block Model

The Stochastic Block Model (SBM), introduced by Holland, Laskey and Leinhardt (1983), is a probabilistic generative model for graphs that assigns nodes to latent blocks and parametrically estimates the connection probabilities between blocks. It is the foundational approach for community detection, core-periphery ident

simulation

Stochastic Cellular Automata

Stochastic Cellular Automata (SCA) extend classical cellular automata by replacing deterministic transition rules with probabilistic ones, allowing each cell on a grid to change state according to a probability distribution conditioned on its neighborhood. This makes SCA a powerful tool for simulating real-world spatia

simulation

Stochastic Differential Equations

Stochastic differential equations (SDEs) are differential equation models that combine a deterministic drift term — governing the average tendency of a system — with a stochastic diffusion term driven by a Wiener process (Brownian motion). Pioneered through Itô calculus by Kiyosi Itô in 1944 and given a comprehensive n

simulation

Stochastic Discrete-Event Simulation

Stochastic Discrete-Event Simulation (Stochastic DES) models complex systems by advancing simulated time from one discrete event to the next, drawing event durations and inter-arrival times from fitted probability distributions. It is the standard technique for analyzing queues, manufacturing lines, healthcare pathways

machine learning

Stochastic Gradient Descent

Stochastic Gradient Descent (SGD) is a first-order iterative optimization algorithm, rooted in the stochastic approximation framework introduced by Robbins and Monro in 1951, that minimizes an objective function by updating model parameters using the gradient computed on a single randomly selected training example (or

simulation

Stochastic Markov Model

A Stochastic Markov Model is a simulation technique that represents a system as a set of mutually exclusive health or decision states, moves a cohort (or individual agents) through those states using probabilistically sampled transition parameters, and aggregates outcomes across thousands of Monte Carlo iterations to p

simulation

Stochastic Microsimulation

Stochastic Microsimulation tracks a large population of individual units — people, households, or firms — through time by applying random draws from empirically estimated probability distributions at each transition event. Unlike deterministic counterparts, every state change is decided by chance, preserving realistic

simulation

Stochastic Multi-Objective Optimization

Stochastic Multi-Objective Optimization (SMOO) is a class of methods that simultaneously optimizes two or more conflicting objectives when parameters, costs, or constraints are uncertain or random. Rather than a single optimal solution, it produces a Pareto front of non-dominated solutions, each representing a differen

simulation

Stochastic Scenario Analysis

Stochastic Scenario Analysis evaluates a system or decision across multiple explicitly defined scenarios, each assigned a probability of occurrence. Unlike deterministic scenario analysis, it propagates uncertainty through probability distributions and computes expected outcomes, variance, and risk metrics across the s

simulation

Stochastic Sensitivity Analysis

Stochastic Sensitivity Analysis (PSA) extends classical one-at-a-time sensitivity testing by representing uncertain model inputs as probability distributions and propagating them through the model via Monte Carlo sampling. The result is a full distribution of possible outputs, together with rankings of which inputs dri

simulation

Stochastic System Dynamics

Stochastic System Dynamics (SSD) extends conventional system dynamics by replacing fixed parameter values and deterministic flow equations with probability distributions and random draws. Running many replications of the stock-flow model yields probabilistic trajectories — confidence bands rather than single lines — en

mining engineering

Stope Layout

Stope layout optimization is the process of designing the size, shape, and spatial arrangement of underground mine excavations (stopes) to maximize ore recovery while maintaining safety and economic viability. It balances the desire for large extraction volumes against rock mechanics constraints and support costs. The

numerical methods

STRIDE/DREAD Threat Modeling

STRIDE/DREAD Threat Modeling is a Microsoft-developed methodology for systematically identifying and prioritizing security threats in software systems. STRIDE enumerates threat categories (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, Elevation of Privilege), and DREAD scores threats by D

astronomy

Strong Gravitational Lensing

Strong gravitational lensing occurs when massive objects (clusters, galaxies) bend light so strongly that multiple images of distant sources appear, or complete rings (Einstein rings) form. Proposed by Sjur Refsdal in 1964 and first observed in 0957+561 in 1979, strong lensing provides direct measurements of lens masse

architecture

Structural Form-Finding

Structural Form-Finding is a computational method for discovering structural geometries that are efficient under given loads and constraints. Pioneered by Heinz Schek in 1974, it reverses traditional structural design: rather than imposing a predetermined form and then analyzing whether it is strong enough, form-findin

civil engineering

Structural Health Monitoring

Structural Health Monitoring (SHM) is a process-based engineering methodology used in civil, mechanical, and aerospace engineering to continuously assess the condition of structures — bridges, buildings, dams, pipelines, and aircraft — through embedded or attached sensor networks. By acquiring real-time or periodic mea

text mining

Structured Text Extraction

Structured text extraction is a document-processing pipeline that automatically identifies and pulls tables, form fields, and structured data from PDF, HTML, and scanned documents. It converts heterogeneous document layouts into machine-readable, analysis-ready records and is widely used in data collection workflows, d

text mining

Subjectivity Detection

Subjectivity detection is a natural-language-processing task that classifies whether a sentence or document conveys objective (neutral information) or subjective (personal opinion, emotion) content. Grounded in the opinion-annotation work of Wiebe and colleagues (2005) and Pang and Lee (2004), it is most often used as

electrical engineering

Subsynchronous Resonance

Subsynchronous Resonance (SSR) is a phenomenon where frequencies below the synchronous frequency (50/60 Hz) are amplified in power systems, causing oscillations that can damage turbines. First observed in Bushland, Texas in 1977, SSR results from interaction between series-compensated transmission lines and synchronous

deep learning

Sundial

Sundial is a family of generative time-series foundation models introduced by Yong Liu and colleagues at Tsinghua University (ICML 2025). Pre-trained on large and diverse time-series corpora, Sundial employs a decomposition-based architecture paired with a generative forecasting head to produce probabilistic multi-hori

astronomy

Sunyaev-Zel'dovich Effect

The Sunyaev-Zel'dovich effect is a phenomenon in which the cosmic microwave background (CMB) is distorted as photons travel through hot gas in galaxy clusters. Proposed by Rashid Sunyaev and Yakov Zel'dovich in 1972, this effect provides a powerful method for detecting distant galaxy clusters and measuring fundamental

machine learning

Support Vector Machine

The Support Vector Machine, introduced by Corinna Cortes and Vladimir Vapnik in 1995, is a classifier that finds the optimal separating hyperplane between classes in a high-dimensional space. It chooses the boundary that leaves the widest possible margin to the nearest training points, which makes its decisions robust

machine learning

Support Vector Regression

Support Vector Regression (SVR), described in Smola and Schölkopf's 2004 tutorial, predicts a continuous outcome by fitting a function that stays within an epsilon-wide tube around the data while incurring as little error as possible. It extends the support vector machine idea from classification to regression, using a

quantum computing

Surface Code Quantum Error Correction

Surface Code is a two-dimensional topological quantum error-correcting code that protects quantum information through geometric redundancy. Introduced by Alexei Kitaev in 2003, surface code is considered the leading candidate for large-scale fault-tolerant quantum computing due to its high error thresholds and feasibil

survey methodology

Survey Weighting

Survey weighting is a statistical procedure that assigns a numeric weight to each sampled unit so that the weighted sample reproduces known population totals. Rooted in classical sampling theory and systematically synthesized by Sharon Lohr (2010), the approach corrects for unequal selection probabilities, unit nonresp

geophysics

SWAT Model

The Soil and Water Assessment Tool (SWAT) is a process-based watershed model that simulates the hydrological cycle, sediment transport, nutrient cycling, pesticide fate, and land management impacts across a watershed or large basin. Developed by Jeff Arnold and colleagues at USDA-ARS in 1998, SWAT has become a standard

deep learning

Swin Transformer

The Swin Transformer is a hierarchical vision transformer introduced by Liu et al. in 2021 that uses shifted window attention to achieve computational efficiency while maintaining strong performance on computer vision tasks. Unlike the original Vision Transformer which applies global self-attention, Swin uses local win

cryptography

Symbolic Execution

Symbolic execution is a program analysis technique that executes programs using symbolic (non-concrete) values instead of actual inputs, tracking how symbolic values flow through the program. Introduced by James C. King in 1976, symbolic execution builds mathematical constraints on program variables and can determine w

cryptography

Symmetric Key Cryptanalysis

Symmetric key cryptanalysis is the study of attacks against symmetric encryption algorithms (such as DES, AES, and stream ciphers) to evaluate their security and identify weaknesses. Classical techniques include differential cryptanalysis and linear cryptanalysis, which have shaped the design of modern ciphers and rema

model evaluation

Symmetric MAPE

Symmetric Mean Absolute Percentage Error is a refinement of MAPE that addresses its asymmetry by using the average of actual and predicted values as the denominator. Proposed by J. Scott Armstrong and refined by Makridakis (1993) and Hyndman & Koehler (2006), sMAPE treats over- and under-predictions symmetrically.

electrical engineering

Symmetrical Components

Symmetrical Components is a mathematical technique for analyzing unbalanced three-phase electrical circuits by decomposing them into balanced component sets. Introduced by Charles Fortescue in 1918, the method transforms the complex analysis of unbalanced systems into simpler balanced equivalent circuits. Symmetrical c

time series

Synchrosqueezing Transform

The synchrosqueezing transform is a time-frequency reassignment technique that sharpens the output of the continuous wavelet transform by concentrating energy along instantaneous frequency ridges. Introduced by Ingrid Daubechies and colleagues in 2011, it addresses the fundamental limitation of the standard wavelet tra

privacy

Synthetic Data Generation

Synthetic data generation is a statistical disclosure limitation technique introduced by Donald Rubin in 1993, in which values in a confidential dataset are replaced by draws from a fitted posterior predictive distribution rather than released directly. The resulting artificial records preserve the joint statistical st

simulation

System Dynamics

System dynamics is a continuous simulation method, developed by Jay W. Forrester at MIT in 1961, that represents a complex system through stocks (accumulations), flows (rates of change), and feedback loops. By expressing these relationships as coupled ordinary differential equations, it reproduces how policies, delays,

linguistics

Systemic Functional Linguistics

Systemic Functional Linguistics (SFL) is a framework for analyzing language developed by Michael Halliday, viewing language as a system of meaning-making choices where speakers select from available options to express meanings. The approach emphasizes the relationship between language form and social context, analyzing