按方法、领域与证据探索科学。

Sequence-to-Sequence Model

The sequence-to-sequence (Seq2Seq) model, introduced by Sutskever, Vinyals and Le and by Cho and colleagues in 2014, is an encoder-decoder neural network that maps a variable-length input sequence to a variable-length output sequence. It is the foundation of machine translation, text summarization, dialogue systems and

machine learning

Sequential Pattern Mining

Sequential Pattern Mining discovers ordered patterns that recur across multiple event sequences in a database. Introduced by Agrawal and Srikant in 1995, it extends association-rule mining to time-ordered transactions. A pattern is frequent when it appears as an ordered subsequence in at least a user-specified fraction

deep learning

SGD with Momentum / Adam Optimizer

Stochastic Gradient Descent (SGD) with momentum and its adaptive descendant Adam are the foundational parameter-update algorithms used to train virtually every modern deep learning model. Momentum SGD was formalised by Polyak (1964) and brought into neural network training by Rumelhart, Hinton, and Williams (1986). Ada

aerospace

SGP4 TLE Propagation

SGP4 (Simplified General Perturbations 4) is a rapid orbital propagation method that predicts satellite position and velocity from Two-Line Element (TLE) sets published by NORAD. Developed in the 1970s, SGP4 accounts for atmospheric drag, gravitational perturbations, and solar radiation pressure using simplified analyt

cryptography

SHA Hash Function

The Secure Hash Algorithm (SHA) is a family of cryptographic hash functions standardized by NIST starting in 1993. SHA functions produce fixed-length digests from arbitrary-length input data, serving as a fundamental building block for digital signatures, message authentication, and data integrity verification across s

telecommunications

Shannon Capacity

Shannon's channel capacity theorem, published in 1948, establishes the maximum rate at which information can be reliably transmitted over a noisy channel. Expressed as C = B log2(1 + S/N) for additive white Gaussian noise (AWGN), it is a fundamental bound in information theory and communications engineering. Shannon pr

machine learning

SHAP

SHAP is a model-explanation method, introduced by Scott Lundberg and Su-In Lee in 2017, that uses Shapley values from cooperative game theory to measure how much each feature contributes to an individual prediction, making the output of black-box machine-learning models interpretable. It supports both global explanatio

quantum computing

Shor's Algorithm

Shor's Algorithm is a polynomial-time quantum algorithm for factoring large integers and computing discrete logarithms, problems believed to be intractable on classical computers. Discovered by Peter Shor in 1994, it demonstrated the potential of quantum computers to break widely used cryptographic systems like RSA, ma

psychometrics

Short form Rasch model

The short form Rasch model applies Rasch measurement theory to abbreviated instrument versions. Rather than using all items from a full scale, researchers select a reduced item set and calibrate it under the Rasch model to verify that the shortened instrument preserves interval-level measurement, adequate person separa

psychometrics

Short-Form IRT

Short-form item response theory applies IRT calibration and scoring to abbreviated or shortened psychological scales. It uses item information functions to guide which items to retain from a full-length instrument, then estimates latent trait scores from the reduced item set while preserving psychometric rigor and link

signal processing

Short-Time Fourier Transform

The Short-Time Fourier Transform (STFT) is a fundamental signal analysis technique that computes the frequency content of a signal as it evolves over time by applying the Fourier transform to short, overlapping windows of the signal. Introduced conceptually by Dennis Gabor in 1946, the STFT provides a time-frequency re

mining engineering

Shrinking Core Model

The Shrinking Core Model, formalized by Szekely, Evans, and Sohn in 1976, describes the kinetics of chemical reactions between solid ore particles and surrounding fluids (leaching solutions, roasting gases). As the reaction proceeds from the particle surface inward, an unreacted core shrinks while products accumulate i

deep learning

Siamese Network

A Siamese network is a deep architecture with two (or more) identical, weight-sharing branches that map inputs into an embedding space where similar inputs land close together and dissimilar ones far apart. Introduced by Bromley, LeCun, and colleagues in 1993 for signature verification and revived by Koch et al. (2015)

cryptography

Side-Channel Analysis

Side-channel analysis is a family of attacks that exploit physical properties of cryptographic implementations (timing, power consumption, electromagnetic emissions, cache behavior) to recover secret keys. Introduced by Paul Kocher in 1996, side-channel attacks have repeatedly broken implementations of theoretically se

computer vision

SIFT Feature Detection

SIFT (Scale-Invariant Feature Transform) is a method for detecting and describing distinctive local features in digital images. Introduced by David Lowe in 1999, SIFT extracts keypoints that remain invariant to scale, rotation, and illumination changes, making it highly robust for image matching and object recognition

signal processing

Signal Denoising

Wavelet signal denoising, introduced by David Donoho in 1995, is a non-parametric technique that removes noise from one-dimensional or multidimensional signals by decomposing them into wavelet coefficients, suppressing small coefficients that likely represent noise via a soft-thresholding operator, and reconstructing a

model evaluation

Silhouette Score

The Silhouette Coefficient, introduced by Peter Rousseeuw in 1987, is a metric that measures how similar an object is to its own cluster compared to other clusters. It ranges from -1 to 1, where values close to 1 indicate well-separated and cohesive clusters, values near 0 suggest overlapping clusters, and negative val

deep learning

SimCLR

SimCLR is a self-supervised learning framework introduced by Chen et al. in 2020 that learns visual representations by contrasting similar and dissimilar views of images. The method applies strong data augmentations to create different views of the same image, then trains an encoder to bring similar views close in repr

research design

Simulation-assisted confirmatory research

Simulation-assisted confirmatory research integrates computational simulation — most commonly Monte Carlo methods — into a hypothesis-driven, confirmatory study design. Before or alongside empirical data collection, the researcher runs simulated data under specified model assumptions to establish expected parameter dis

experimental design

Simulation-assisted control chart

Simulation-assisted control chart integrates Monte Carlo or discrete-event simulation with traditional Shewhart-type control charting to design, validate, and optimize chart parameters before deployment on a real process. Rather than relying solely on assumed distributional forms, the practitioner builds a simulation m

experimental design

Simulation-assisted event tree analysis

Simulation-assisted event tree analysis (ETA) extends classical event tree analysis by replacing fixed point-estimate branch probabilities with Monte Carlo or discrete-event simulation. This allows analysts to propagate uncertainty through every branch of the tree and obtain full probability distributions over accident

experimental design

Simulation-assisted failure mode and effects analysis

Simulation-assisted FMEA enhances the classical Failure Mode and Effects Analysis by replacing point-estimate occurrence ratings with probabilistic simulation — typically Monte Carlo — to quantify failure probability distributions across a system's components. This yields statistically grounded Risk Priority Numbers (R

experimental design

Simulation-assisted fault tree analysis

Simulation-assisted fault tree analysis (SA-FTA) combines the logical structure of classical fault tree analysis with Monte Carlo or discrete-event simulation to estimate the probability and timing of an undesired top event when component failures follow complex, non-exponential, or correlated probability distributions

research design

Simulation-assisted hypothesis testing research

Simulation-assisted hypothesis testing research replaces or supplements analytical probability theory with computational simulation — resampling, permutation, or Monte Carlo methods — to construct null distributions and evaluate hypotheses. Rather than assuming a parametric distribution and consulting a table, the rese

experimental design

Simulation-assisted process capability analysis

Simulation-assisted process capability analysis combines Monte Carlo simulation with classical capability indices (Cp, Cpk, Cpm) to evaluate whether a process can consistently meet specification limits when direct measurement is costly, dangerous, or impractical. By propagating input distributions through a process mod

research design

Simulation-assisted quantitative content analysis

Simulation-assisted quantitative content analysis (SA-QCA) extends classical quantitative content analysis by integrating computational simulation — typically Monte Carlo methods or agent-based models — to validate coding schemes, estimate coder reliability under controlled conditions, test category distinctiveness, an

experimental design

Simulation-assisted reliability analysis

Simulation-assisted reliability analysis combines probabilistic reliability theory with computational simulation — most commonly Monte Carlo methods or finite-element models — to estimate the probability that a system, component, or structure will perform its intended function under uncertain operating conditions. Rath

experimental design

Simulation-assisted statistical process control

Simulation-assisted statistical process control (SA-SPC) combines computer simulation — typically Monte Carlo or discrete-event simulation — with classical SPC methods to design, test, and calibrate control charts and monitoring schemes before or alongside deployment on a real production process. Rather than relying so

research design

Simulation-Assisted Trend Research

Simulation-assisted trend research combines repeated cross-sectional survey data collected at multiple time points with computational simulation techniques — such as Monte Carlo methods or agent-based modeling — to project, validate, and stress-test observed trends. It extends classic trend research by replacing or sup

control theory

Simultaneous Localization and Mapping

Simultaneous Localization and Mapping (SLAM) is the problem of enabling a mobile robot to build a map of its environment while simultaneously determining its own location within that map using noisy sensor measurements. Formulated by Durrant-Whyte and Bailey in 2006, SLAM is fundamental to autonomous robotics, enabling

bioinformatics

Single-cell ChIP-seq peak calling

Single-cell ChIP-seq peak calling is a bioinformatics pipeline that identifies genomic regions enriched for histone modifications or transcription factor binding in individual cells. By profiling chromatin states at single-cell resolution, it reveals epigenomic heterogeneity hidden in bulk ChIP-seq experiments, enablin

bioinformatics

Single-cell Copy Number Variation Analysis

Single-cell copy number variation (scCNV) analysis detects gains and losses of genomic segments within individual cells, enabling researchers to resolve intratumor heterogeneity, reconstruct clonal evolution, and distinguish malignant from normal cells at single-cell resolution. It can be applied to single-cell whole-g

bioinformatics

Single-cell epigenome-wide association study

A single-cell epigenome-wide association study (scEWAS) interrogates epigenetic marks — primarily DNA methylation or chromatin accessibility — across the entire genome at single-cell resolution, then statistically associates variation in those marks with a phenotype, disease, or exposure. By resolving cell-type heterog

bioinformatics

Single-cell eQTL analysis

Single-cell eQTL analysis identifies genetic variants (eQTLs) that regulate gene expression in a cell-type-specific manner by jointly analysing single-cell RNA-seq profiles and donor genotype data. Unlike bulk eQTL methods, it resolves regulatory effects that are diluted or masked when cell types are mixed, enabling di

bioinformatics

Single-cell Gene Set Enrichment Analysis

Single-cell gene set enrichment analysis (scGSEA) extends classical bulk GSEA to the resolution of individual cells. Rather than testing whether a gene set is enriched in a sample-level comparison, scGSEA assigns an enrichment or activity score to each cell, enabling researchers to map pathway activity across heterogen

bioinformatics

Single-cell GWAS

Single-cell GWAS is an integrative bioinformatics pipeline that maps genome-wide association study (GWAS) signals onto single-cell transcriptomic landscapes to identify which cell types and individual cells carry disproportionate genetic risk for a disease or trait. By leveraging single-cell RNA-seq atlases alongside G

bioinformatics

Single-cell metabolomics analysis

Single-cell metabolomics analysis measures the small-molecule metabolite content of individual cells, revealing cell-to-cell metabolic heterogeneity that bulk methods obscure by averaging. Rooted in mass spectrometry and microfluidics advances, it enables researchers to map metabolic states across cell populations, ide

bioinformatics

Single-cell Microbiome Diversity Analysis

Single-cell microbiome diversity analysis resolves the composition and functional heterogeneity of microbial communities at the level of individual cells or bacteria. By combining single-cell or single-bacterium isolation with high-throughput sequencing, this pipeline overcomes the averaging effect of bulk metagenomics

bioinformatics

Single-cell Phylogenetic Analysis

Single-cell phylogenetic analysis reconstructs evolutionary or developmental trees from single-cell sequencing data, tracing how individual cells diverged from a common ancestor. By leveraging somatic mutations, CRISPR-introduced barcodes, or copy-number changes as heritable characters, this method maps clonal relation

bioinformatics

Single-cell RNA-seq analysis

Single-cell RNA sequencing (scRNA-seq) analysis characterises gene expression at the resolution of individual cells, enabling discovery of cell types, states, and transitions that are invisible in bulk transcriptomics. Starting from raw sequencing reads, the workflow produces a cell-by-gene count matrix and proceeds th

bioinformatics

Single-cell RNA-seq differential expression

Single-cell RNA-seq differential expression (scRNA-seq DE) analysis identifies genes whose expression levels differ significantly between defined groups of individual cells — such as cell types, disease states, or treatment conditions. Unlike bulk RNA-seq, which averages signals across millions of cells, scRNA-seq DE o

bioinformatics

Single-cell sequence alignment

Single-cell sequence alignment is the computational step that maps millions of short sequencing reads produced by single-cell RNA-seq experiments back to a reference genome or transcriptome. Unlike bulk RNA-seq alignment, each read carries a cell barcode and a Unique Molecular Identifier (UMI) that together identify th

bioinformatics

Single-cell variant calling

Single-cell variant calling is a bioinformatics pipeline that identifies DNA sequence variants — single-nucleotide variants (SNVs), small insertions and deletions, and copy-number alterations — within individual cells rather than across a bulk tissue mixture. By resolving the mutational landscape cell by cell, it revea

numerical methods

Singular Value Decomposition

Singular Value Decomposition (SVD) is a fundamental matrix factorization technique that decomposes any m × n matrix A into the product A = U Σ V^T, where U and V are orthogonal matrices and Σ is a diagonal matrix of singular values. Developed by Gene Golub and others in the 1960s–1970s, SVD is the most robust method fo

meteorology

Skew-T Log-P Analysis

The Skew-T Log-P diagram is a thermodynamic chart used extensively in meteorology to visualize atmospheric profiles of temperature, dew point, and pressure. Developed in its modern form by Reitan in the 1960s, it allows forecasters and researchers to quickly assess atmospheric stability, convective potential, wind shea

mining engineering

Slag Basicity

Slag basicity is a measure of the composition of slag formed during smelting and roasting operations. It is typically expressed as the ratio of basic oxides (CaO, MgO) to acidic oxides (SiO2). Basicity controls slag fluidity, viscosity, and reactivity, directly affecting metal recovery, processing temperature, and prod

control theory

Sliding Mode Control

Sliding Mode Control (SMC) is a robust nonlinear control technique that forces a system to follow a predetermined surface (the sliding surface) in state space by using discontinuous (bang-bang or high-frequency switching) control inputs. Developed by Utkin and further advanced by Slotine, SMC is remarkably insensitive

civil engineering

Slope Stability (Bishop-Janbu)

The Bishop and Janbu methods are limit equilibrium approaches for analyzing slope stability, computing the factor of safety against shear failure along a potential slip surface. Developed by Bishop (1955) and Janbu (1954), these methods remain the most widely used tools in geotechnical engineering for evaluating cut sl

text mining

Slot Filling

Slot filling is a natural-language-understanding task that extracts predefined template fields — such as date, location, or product name — from a user utterance. It emerged as a core component of dialogue systems and form-based information extraction, and became widely studied after Goo et al. (2018) introduced the Slo

telecommunications

Slotted ALOHA

Slotted ALOHA is a fundamental random access protocol enabling multiple devices to share a wireless channel without centralized coordination. Introduced by Abramson (1970) and refined by Roberts (1975), it divides time into fixed slots and allows devices to transmit at the beginning of a slot with a fixed probability.

network analysis

Small-World and Scale-Free Network Analysis

Small-world and scale-free network analysis tests whether a real-world network exhibits two landmark topological signatures identified in 1998-1999: the Watts-Strogatz small-world property (high local clustering combined with short average path lengths) and the Barabási-Albert scale-free property (a degree distribution

electrical engineering

Smart Grid State Estimation

Power system state estimation infers the real-time voltage and phase angle at every bus in a power network from redundant measurements of power flows and voltages. It is the foundation of modern grid operations, enabling real-time monitoring, contingency analysis, and optimal control. Advanced state estimation with syn

electrical engineering

Smith Chart

The Smith Chart is a graphical tool for visualizing and manipulating complex impedances and reflection coefficients on transmission lines. Introduced by Phillip Smith in 1939, the chart maps the complex reflection coefficient plane to a circular chart, enabling intuitive graphical analysis of transmission line problems

fluid dynamics

Smoothed Particle Hydrodynamics

Smoothed Particle Hydrodynamics (SPH) is a meshfree particle method for simulating fluid dynamics, developed independently by Lucy in 1977 and Gingold and Monaghan in 1977. Rather than discretizing on a fixed grid, SPH represents fluids as collections of particles that carry mass, momentum, and energy. Each particle in

text mining

Social Media NLP

Social Media NLP is a specialised natural-language-processing pipeline designed for the short, noisy, and informal text that appears on platforms such as Twitter, Reddit, and comment sections. Unlike general-purpose NLP, this pipeline accounts for platform-specific conventions — hashtags, emojis, abbreviations, and cod

network analysis

Social Network Analysis

Social Network Analysis (SNA) is a structural method that maps and measures relationships and flows between people, groups, organizations, or other entities modeled as nodes connected by ties (edges). Rather than focusing on individual attributes, SNA reveals how the pattern of connections shapes behavior, influence, i

software engineering

Software Complexity Metrics

Software complexity metrics quantify the structural and operational difficulty of code through numerical measurements. Introduced by Thomas McCabe in 1976, cyclomatic complexity became the foundational approach. These metrics assess maintainability, testability, and defect risk, enabling teams to identify problematic c

civil engineering

Soil-Structure Interaction

Soil-structure interaction (SSI) analysis accounts for the dynamic coupling between a structure and its supporting foundation soil, recognizing that the soil is not infinitely rigid. Formalized by Veletsos in 1974, this approach reveals how foundation compliance, radiation damping, and kinematic effects modify the stru

acoustics

Sonar Equation

The sonar equation is a fundamental framework for predicting the detection range and performance of active and passive sonar systems in underwater environments. Systematized by Robert Urick in his seminal 1983 work, the sonar equation quantifies the acoustic signal-to-noise ratio (SNR) needed for detection, accounting

acoustics

Sound Transmission Class

Sound Transmission Class (STC) is a single-number rating used to describe how well building elements (walls, doors, windows) reduce sound transmission between adjacent spaces. Standardized by ASTM International and ISO, STC is calculated from sound transmission loss (STL) measurements across the speech frequency range