按方法、领域与证据探索科学。

Optimality Theory

Optimality Theory (OT) is a constraint-based framework for modeling phonology and syntax, developed by Alan Prince and Paul Smolensky in 1993. The core idea is that languages produce the optimal output that best satisfies a ranked hierarchy of universal constraints. Rather than listing rules, OT explains linguistic phe

text mining

Abbreviation Expansion

Abbreviation and acronym resolution is a natural-language-processing pipeline that maps each short form in a text to its full-length definition using contextual cues from the surrounding text. It is especially important in medical, legal, and technical documents, where the same acronym may carry entirely different mean

model evaluation

Accuracy

Accuracy is the proportion of correct predictions among the total number of predictions made by a classification model. It is the most intuitive performance metric and measures how often the classifier makes correct predictions overall, regardless of class.

machine learning

Active learning Association rules

Active learning association rules combines the iterative query-and-label loop of active learning with association rule mining, allowing a human expert to guide the discovery process interactively. Instead of exhaustively enumerating all rules above a fixed support-confidence threshold, the system selects the most infor

machine learning

Active Learning Autoencoder Anomaly Detection

Active Learning Autoencoder Anomaly Detection combines an autoencoder's unsupervised reconstruction-error scoring with an active learning query loop. The model flags high-error instances as candidate anomalies, selectively asks a human oracle to label the most informative ones, and iteratively retrains — achieving stro

machine learning

Active learning Gaussian mixture model

Active Learning Gaussian Mixture Model combines an iterative query strategy with a Gaussian Mixture Model learner. The algorithm selects the most informative unlabeled points — typically those with highest predictive uncertainty — presents them to an oracle for labeling, and refits the GMM using EM on the growing label

simulation

Agent-based cellular automata

Agent-Based Cellular Automata (ABCA) is a hybrid simulation framework that integrates the local transition rules of cellular automata with the autonomous behavioral logic of agent-based modeling. Cells in a spatial grid both evolve according to neighborhood rules and host agents that perceive, decide, and act, enabling

numerical methods

ATAM

The Architecture Tradeoff Analysis Method (ATAM) is a systematic technique developed by Kazman et al. at CMU/SEI for evaluating software architectures against quality attributes (performance, security, modifiability). ATAM uncovers architectural risks and tradeoffs early, helping teams assess whether designs meet quali

deep learning

BERT-based Classification

BERT-based Classification fine-tunes Google's Bidirectional Encoder Representations from Transformers model on a labelled text dataset, replacing the generic pre-trained head with a task-specific classification layer. It exploits deep bidirectional context from hundreds of millions of pre-trained parameters to deliver

biomechanics

CFD Hemodynamics

Computational fluid dynamics (CFD) for hemodynamics solves the Navier-Stokes equations to simulate blood flow in realistic vascular geometries. Pioneered by researchers such as David Steinman, CFD hemodynamics reveals complex flow patterns, wall shear stress distributions, and hemodynamic factors implicated in atherosc

numerical methods

CK Metrics

CK Metrics is a suite of six object-oriented design metrics introduced by Chidamber and Kemerer in 1994 to measure class complexity, cohesion, and coupling. The metrics quantify OO design quality; high coupling and low cohesion predict defects and maintenance effort.

manufacturing

CNC Tool Path Generation

CNC tool path generation is the computational process of determining the precise sequence and trajectory of tool movements required to machine a workpiece on computer numerical control (CNC) machines. Developed from the intersection of numerical control automation and computational geometry in the 1990s, this method tr

machine learning

Conformal Prediction

Conformal Prediction is a distribution-free framework for constructing statistically valid prediction sets (for classification) or prediction intervals (for regression) around the output of any pre-trained machine learning model. Introduced by Vovk, Gammerman, and Shafer in their 2005 monograph, it provides a finite-sa

nuclear physics

Criticality Safety Analysis

Criticality safety analysis is a systematic evaluation of fissile material systems to ensure nuclear chain reactions remain controlled, originating from Hahn and Strassmann's 1938 discovery of nuclear fission. It determines safe limits on fissile mass, concentration, geometry, and spacing using neutron transport calcul

deep learning

CycleGAN

CycleGAN, introduced by Zhu et al. at ICCV 2017, learns to translate images between two visual domains without requiring paired training examples. It trains two generators and two discriminators simultaneously, enforcing a cycle-consistency constraint so that an image translated from domain X to Y and back again recove

deep learning

Domain-adaptive BERT-based Classification

Domain-adaptive BERT-based classification extends the standard fine-tuning pipeline by first continuing BERT's masked-language-model pre-training on a large corpus of in-domain unlabeled text, then fine-tuning the adapted model on labeled examples for the target classification task. This two-stage approach closes the v

deep learning

Domain-adaptive diffusion model

A domain-adaptive diffusion model is a denoising diffusion probabilistic model (DDPM) that is pre-trained on large general datasets and then adapted — through fine-tuning, textual inversion, or LoRA — to generate high-quality outputs in a specific target domain. It combines the powerful generative capacity of diffusion

deep learning

Domain-adaptive GRU

Domain-Adaptive GRU combines the Gated Recurrent Unit architecture with domain adaptation techniques to train a sequence model on a labeled source domain and transfer it to a different but related target domain, reducing performance degradation caused by distribution shift. It is widely applied in NLP tasks such as cro

deep learning

Domain-adaptive variational autoencoder

A Domain-Adaptive Variational Autoencoder (DA-VAE) extends the standard VAE framework to learn disentangled latent representations that separate domain-specific variation from class-relevant and domain-invariant content, enabling models trained on a source domain to generalise effectively to a different but related tar

machine learning

Emerging Pattern Mining

Emerging Pattern Mining (EPM) is a contrast-based data mining technique that identifies itemsets whose support increases significantly — or jumps from zero — when moving from one dataset (or class) to another. Introduced by Dong and Li in 1999, it is primarily used in classification, anomaly detection, and trend analys

materials science

Finite Element Analysis

Finite Element Analysis (FEA) is a numerical technique for obtaining approximate solutions to boundary value problems described by differential equations. Developed systematically by Richard Courant in 1943 and popularized by Clough in the 1960s, FEA divides a complex domain into smaller, simpler elements to solve engi

applied physics

Gravity Assist

A gravity assist (or swing-by) maneuver uses the gravitational field of a planet or other celestial body to alter a spacecraft's trajectory and velocity without expending fuel. Discovered by Michael Minovitch at JPL in 1961, this technique is crucial for reaching distant planets economically. It works by exploiting the

bioinformatics

Machine learning-assisted gene set enrichment analysis

Machine learning-assisted gene set enrichment analysis (ML-GSEA) extends the classical GSEA framework by incorporating supervised or unsupervised ML models — such as random forests, neural networks, or deep learning architectures — to improve the detection, ranking, and biological interpretation of enriched gene sets f

linguistics

N400/P600 Analysis

N400/P600 Analysis is a neurocognitive method using electroencephalography (EEG) to measure event-related potentials (ERPs) that reflect brain responses to linguistic stimuli. The N400 component (a negative deflection at 400 ms) indexes semantic processing and surprise; the P600 component (a positive deflection at 600

manufacturing

Tolerance Stack-up

Tolerance stack-up analysis is a method for predicting the cumulative effect of manufacturing tolerances on the final dimensions and fit of assembled components. When parts with individual tolerances are assembled together, their tolerances combine in complex ways, often producing a result that is worse than each part

architecture

Acoustic Design Analysis

Acoustic Design Analysis is a method for evaluating the acoustical properties of buildings to predict sound levels, reverberation time, and speech intelligibility. Founded by Wallace Clement Sabine in the early 1900s, the field encompasses room acoustic design (controlling reverberation), sound transmission loss (preve

oceanography

Acoustic Doppler Current Profiler

The Acoustic Doppler Current Profiler (ADCP) is an instrument that uses Doppler-shifted acoustic backscatter to measure water velocity profiles along a vertical profile. Developed by RD Instruments in the 1980s, it has become the standard method for high-resolution current profiling in oceanographic research. ADCPs pro

linguistics

Acoustic Phonetics

Acoustic Phonetics is the study of the physical properties of speech sounds using instrumentation to measure and analyze sound waves. Pioneered by Peter Ladefoged and Kenneth Stevens, this method uses spectrograms, formant analysis, and waveform measurements to characterize vowels, consonants, and prosodic features wit

acoustics

Acoustic Ray Tracing

Acoustic ray tracing is a computational technique for predicting sound propagation in rooms by treating acoustic energy as rays that reflect specularly off surfaces. Formalized by Allen and Berkley in 1979 via the image source method, ray tracing is one of the most computationally efficient methods for room acoustic si

control theory

Active Disturbance Rejection Control

Active Disturbance Rejection Control (ADRC) is a control method that estimates and cancels disturbances and model uncertainties in real-time using an extended state observer (ESO), treating them as additional 'disturbance states'. Developed by Han and popularized by Gao, ADRC achieves remarkable robustness without requ

machine learning

Active Learning

Active learning is an iterative machine-learning paradigm in which a learning algorithm selectively queries an oracle — typically a human annotator — for labels on the most informative unlabeled examples. Formalized by Burr Settles in his seminal 2009 literature survey, active learning addresses the practical bottlenec

machine learning

Active learning Boosting

Active Learning Boosting combines the query-driven label acquisition of active learning with the weighted-ensemble logic of boosting algorithms such as AdaBoost. The model iteratively selects the most informative unlabeled examples to annotate — guided by the disagreement or uncertainty within the boosting ensemble — a

machine learning

Active learning Decision tree

Active learning with a decision tree combines the interpretable structure of a CART-style tree with a query strategy that selects the most informative unlabeled instances for human annotation. The model iteratively requests labels only for examples it is most uncertain about, minimising labeling cost while maximising c

machine learning

Active Learning Federated Learning

Federated Active Learning combines the annotation-efficiency of active learning with the privacy-preserving decentralization of federated learning. A shared global model is trained across distributed clients, each of which independently ranks its unlabeled local data and requests labels only for the most informative ex

machine learning

Active Learning Gradient Boosting

Active Learning Gradient Boosting combines the powerful predictive accuracy of gradient boosted trees with an active learning loop that selects the most informative unlabeled examples for human annotation. By querying only the instances the model is most uncertain about, the method achieves high accuracy with far fewer

machine learning

Active learning Isolation forest

Active Learning Isolation Forest combines the unsupervised anomaly-scoring power of Isolation Forest with an iterative query strategy that asks a human expert to label the most informative instances. The result is a detector that refines its anomaly boundaries using a minimal labeling budget, dramatically improving pre

machine learning

Active learning K-nearest neighbors

Active learning with K-nearest neighbors combines the instance-based prediction of KNN with an iterative query strategy that selects the most informative unlabeled examples for annotation. The model requests labels only for instances where neighborhood vote margins are narrowest, achieving competitive accuracy with far

machine learning

Active Learning LightGBM

Active Learning LightGBM couples the query-efficient label-selection strategy of active learning with the speed and accuracy of LightGBM, a histogram-based gradient boosting framework. The model iteratively selects the most informative unlabeled instances for human annotation, retrains LightGBM on the growing labeled s

machine learning

Active Learning Linear Regression

Active Learning Linear Regression is an iterative machine-learning approach that couples a linear regression model with an intelligent query strategy to select the most informative unlabeled points for labeling. By focusing labeling effort where uncertainty is highest, it achieves competitive predictive accuracy with f

machine learning

Active learning One-class SVM

Active Learning One-class SVM combines the one-class support vector machine — a kernel-based novelty detector that learns the boundary of normal data — with an active learning loop that selects the most informative unlabeled instances for expert annotation. The result is a data-efficient anomaly detector that improves

machine learning

Active Learning Self-supervised Learning

Active learning combined with self-supervised learning leverages unlabeled data through self-supervised pre-training to build rich representations, then uses an active query strategy to select the most informative examples for human annotation, maximizing model performance under a tight labeling budget. This hybrid app

machine learning

Active learning Stacking ensemble

Active Learning Stacking Ensemble combines an active learning query loop with stacked generalization: a pool of unlabeled data is available, and the model iteratively selects the most informative instances for human labeling, using those labels to train and refine a stacking ensemble of multiple base learners topped by

machine learning

Active learning Support vector machine

Active learning SVM combines the strong decision-boundary of support vector machines with an intelligent query strategy that selects the most informative unlabeled instances for human annotation. Introduced by Tong and Koller in 2001, it achieves high classification accuracy using far fewer labeled examples than passiv

machine learning

Active Learning Voting Ensemble

Active Learning Voting Ensemble — formally known as Query by Committee — is an active learning strategy that trains a committee of diverse models and selects the unlabeled examples where the committee members disagree most for human annotation. By focusing labeling effort on the most informative points, it achieves hig

machine learning

AdaBoost

AdaBoost (Adaptive Boosting) is the original boosting algorithm, introduced by Yoav Freund and Robert Schapire in 1997, that combines a sequence of simple weak learners by giving more weight to the observations they get wrong. The forerunner of gradient boosting, it is simple, interpretable, and a strong baseline for c

control theory

Adaptive Control

Adaptive Control is a control strategy that adjusts controller parameters in real-time based on online system identification to maintain performance despite changing plant dynamics or uncertain parameters. Pioneered by Astrom and Wittenmark, adaptive control enables robust operation in time-varying environments, from a

signal processing

Adaptive LMS Filter

The Least Mean Squares (LMS) filter is an adaptive signal processing algorithm that continuously updates filter coefficients to minimize the squared error between the filter output and a desired signal. Introduced by Bernard Widrow and Marcian Hoff in 1960, the LMS algorithm is one of the most widely used adaptive filt

manufacturing

Additive Manufacturing Slicing

Additive manufacturing slicing is the computational process of converting a three-dimensional CAD model into a series of two-dimensional cross-sectional layers that are sequentially built up by 3D printing hardware. Developed during the early maturation of stereolithography and selective laser sintering in the 1990s, t

model evaluation

Adjusted R-squared

Adjusted R² is a corrected version of the coefficient of determination that accounts for the number of predictors in a regression model. Introduced by Henri Theil in 1961, it addresses the fundamental limitation of standard R²: the tendency to increase whenever any predictor is added, regardless of whether that predict

model evaluation

Adjusted Rand Index

The Adjusted Rand Index (ARI), developed by Hubert and Arabie in 1985, is an external clustering evaluation metric that measures the agreement between a predicted clustering and a ground truth labeling. It ranges from -1 to 1, where 1 indicates perfect agreement, 0 indicates random clustering, and negative values indic

genetics

Admixture Analysis

Admixture analysis is a population genetics method that infers population structure and individual ancestry from multilocus genotype data. Originally developed by Pritchard, Stephens, and Donnelly (2000) and refined by Alexander, Novembre, and Lange (2009), admixture analysis reveals how genetic variation is distribute

applied physics

Adsorption Isotherm (Langmuir-Freundlich)

Adsorption isotherms describe the equilibrium uptake of a substance on a solid surface as a function of gas or solution phase concentration at constant temperature. The Langmuir isotherm (1918) and Freundlich isotherm (1906) are classical empirical models. The Langmuir model assumes monolayer coverage and is mechanisti

deep learning

Adversarial Training

Adversarial Training is a robust optimization procedure for deep neural networks in which the model is trained not on clean data alone but on worst-case perturbed inputs crafted during training. Formalized by Madry et al. (2018) as a min-max saddle-point problem, the method uses Projected Gradient Descent (PGD) to gene

geophysics

Aerosol Optical Depth

Aerosol Optical Depth (AOD) is a dimensionless measure of aerosol light extinction in the atmosphere, quantifying how much sunlight is scattered and absorbed by particles suspended in air. Formalized by Ångström in 1929 and now routinely measured via satellite (MODIS, Sentinel-5P) and ground networks (AERONET), AOD is

cryptography

AES (Rijndael)

The Advanced Encryption Standard (AES), also known as Rijndael, is a symmetric block cipher adopted as the official encryption standard by the U.S. government in 2001. It processes data in 128-bit blocks using 128, 192, or 256-bit keys and performs multiple rounds of substitution, permutation, and mixing operations. AE

machine learning

Affinity Propagation

Affinity propagation, introduced by Brendan Frey and Delbert Dueck in 2007, is a clustering algorithm that identifies representative 'exemplars' among the data by exchanging messages between every pair of points until a consistent set of clusters emerges. Unlike k-means it does not require the number of clusters to be

simulation

Agent-based Discrete-Event Simulation

Agent-based discrete-event simulation (AB-DES) is a hybrid modeling paradigm that couples autonomous agent behavior with an event-driven execution engine. It captures the decision-making heterogeneity of individual entities while maintaining the precise, time-stamped flow control of discrete-event simulation, making it

simulation

Agent-based Markov model

The Agent-Based Markov Model (ABMM) is a hybrid simulation framework that embeds Markov chain state-transition logic inside individual autonomous agents. Each agent independently samples its next state from a probability transition matrix, enabling the model to capture both micro-level heterogeneity across agents and t

simulation

Agent-based microsimulation

Agent-based microsimulation (ABMS) merges traditional microsimulation's individual-level statistical tracking with agent-based modeling's behavioral rules and interaction mechanisms. It creates virtual populations of heterogeneous agents who evolve over time according to transition probabilities, adaptive behaviors, an

simulation

Agent-Based Modeling

Agent-based modeling (ABM) is a computational simulation method, formalized through the work of Thomas Schelling and Robert Axelrod in the 1970s–1990s, that simulates the behavior of complex systems by specifying and running autonomous agents — individuals, firms, cells, or any bounded entity — whose local interactions