List of algorithms
A complete list of all major algorithms (300), across the board. In order to provide a ready-made program for each, or a description of the algorithm. Programming languages include Java, JavaScript and PHP, C or C++ either directly or generated from a source in Script.
- Automatic machine
- Bioinformatics and chemoinformatics
- Reduction
- Cryptography
- Random number generator
- Software engineering
- Memory allocation
- Distributed systems
- Operating system (disk, schedule...)
- Geometry
- Columns
- Graphics
- Artificial intelligence
- Lists, Tables, and Trees
- Mathematics
- Bases
- Optics
- Optimization and online search
- Parsing
- Forecasting (statistics)
- Logic programming
- Part
- Sciences
- (Processing) Signal
- Text
- Utilities
- Misc
Automatic machine
- Powerset builds. Algorithm for converting a nondeterministic automaton to a deterministic one.
- Todd-Coxeter algorithm. To create cosets:
Bioinformatics and chemoinformatics
- Nedleman-Wunsch. Performs global alignment on two sequences, for proteins or nucleotides.
- Smith-Waterman. Nedleman-Wunsch variation.
- Ullmann's algorithm for solving the subgraph isomorphism problem (isomorphism subgraph) of solving. (1976). Defines two graphs having an isomorphic subgraph.
The problem of largest isomorphic subgraphs is solved by a modular compositional graph.
Reduction
Lossless compression
- Burrows-Wheeler transform. Preprocessing to improve compression.
- Deflate. Data compression, using ZIP.
- Delta encodes. Help compress data in which sequences often appear.
- Incremental coding. Delta encoding applied to string sequences.
- LZW. (Lempel-Ziv-Welch). Successor to LZ78. Create a translation table from the data to be compressed. Used by GIF graphic format.
- LZ77 and 78. The basis of the upcoming declensions in LZ (LZW, LZSS,...). They all use dictionary encoding.
- LZMA. Initials of the Lempel-Ziv-Markov-Algorithm circuit.
- LZO. Speed-oriented algorithm.
- PROMILLE. (Partial coincidence prediction). Partial absentee prediction. Adaptive and statistical compression technique.
- Shannon-Fano encodes. Builds set-top boxes based on character set and their probabilities.
- Truncated binary encoding. The encoding commonly used for uniform distribution with the final alphabet. Improves binary encoding.
- Rune-length encodes. A primary algorithm that replaces a single sequence with the number of instances.
- Sequitur. Uses extra grammar for the string.
- EZW (Embedded Zerotree Wavelet). Progressive coding to compress an image into a set of bits with progressive accuracy. Can be used in lossy compression for a more effective result.
- Zopfli. Google, effective but slow compression. The name means small bread in Swiss German.
- Brotley. Google, the successor to Zopfli, based on LZ 77, encodes entropy and contextual modeling .
- Huffman encodes. Uses the relative frequency of characters.
- Adaptive Hufman encodes. Adaptive coding technique based on Huffman algorithm.
- Arithmetic coding. The encoding is developed.
- Range encoding. Even princely, like arithmetic coding, but with a different method.
- No encoding. Represents the number n with n others followed by zero.
- Elias delta, gamma, omega coding. Universal encoding of positive integers.
- Fibonacci coding. Encoding positive integers into binary words.
- Golomb encodes. Optimal coding of alphabets by geometric distribution.
- Rice encodes. Like Golomb.
Entropy coding
A coding principle that assigns codes to symbols so that the length of the codes corresponds to the probability of the symbols.
Lossy Compression
- Linear predictive coding. Uses a spectral envelope in the compressed form of a digital voice signal.
- Legitimate algorithm.
- Mu-law algorithm. Analog signal compression.
- Fractal compression. Compressing images using fractals.
- Encoding conversion. Used for sound or photo type data.
- Vector quantization. Method used in lossy compression.
- Wawelet compression. The compression type is suitable for picture and sound.
Cryptography
Private key encryption (symmetric)
Uses a secret key (or a pair of directly related keys) for both encryption and decryption.- Advanced Encryption Standard (AES), also known as Rijndael.
- Blaufisch. Developed by Schneir as a general use algorithm to replace aging DE.
- Data Encryption Standard (DES). Formerly DE Algorithm.
- IDEA (International Data Encryption Algorithm). Former IPES (Improved PES), also replaces DES. PGP (Pretty Good Privacy) is used. Converts chunked data with a key.
- RC4 (or ARC4). Stream encryption is widely used, including SSL for Internet traffic and WEP for wireless networks.
- Tini encryption algorithm. An easy-to-implement data block algorithm using multiple formulas.
- PES (proposed encryption standard). Previous name IDEA.
Public key encryption (asymmetric)
Uses a key pair, the so-called public and private key. The public key encrypts the message, only the private key allows it to be decrypted.- DSA (digital signature algorithm). Generates keys with a prime and a random number. Used by US agencies, and now in the public domain.
- ElGamal. Based on Difie-Hellman, used in GNU Privacy Guard, PGP and other cryptographic systems.
- RSA (Rivest, Shamir, Adleman). Widely used in e-commerce. Uses primes.
- Diffie-Hellman (Merkle) key exchange (or exponential exchange key). Method and algorithm for exchanging secret content over an unsecured communication channel. RSA is used.
- NTRUEncrypt. Uses polynomial rings with convolutional multiplications.
Control code generator
Generates code from an encrypted message or file of any size.- MD5. Used to test CD or DVD ISO images.
- RIPEMD (RACE Integrity Primitives Evaluation Message Digest). It is based on the principles of MD4 and is similar to ShA-1 .
- SHA-1 (Secure Hash Algorithm 1). Most used in all cryptographic hashes of SHA functions. Developed by the US agency NSA.
- HMAC. Authentication of messages by hash keys.
- Tiger (TTH). Used in the Tiger Tree hash.
Secure encryption using random numbers
- Look. Random number generator
Cryptography methods
Secret Exchange, Secret Splitting, Key Splitting, M.N.- Shamir's secret scheme. It is a formula based on polynomial interpolation.
- Blackley's secret exchange scheme. This is geometry, the secret is a point in m-dimensional space.
Other methods and interpretation
- Shore's algorithm. A putative quantum algorithm capable of decrypting code based on asymmetric functions such as RSA.
- Substrates sum. A set of integers, given whether there is a subset whose sum is zero? Used in cryptography.
Random number generators
- Bloom Bloom Shub. Based on a prime formula.
- Mersenne twister. Matsumoto Nishimura, fast, has a very long periodicity.
- Lagged Fibonacci generator. Improves the linear congruence generator using the Fiboslavci sequence.
- Linear Congregator. One of the oldest, not the best, generates a sequence of three numbers.
- yarrow algorithm. Bruce Schneier, John Kelsey and Nils Ferguson. A cryptographically secure generator can be used to generate truly random numbers from the inputs of analog devices.
- Fortune. Presented as an improvement to the Yarrow algorithm.
- Linear feedback shift register. An offset register whose input is a linear function of the previous content. The original content is "seed" (grain).
Software engineering
- Semantics repair and isolation algorithms.
- Insert Unicode. Provides a standard way to arrange names, words, or strings in a specific sequence.
- CHS conversion. Conversion between disk address systems.
- Cyclic redundancy check. Calculation of control words.
- Parity control. Elementary error detection method. Is the number even or odd?
Memory allocation
- Garage collector. Memory manager.
- Buddy's Memory Manual. Allocate memory by reducing fragmentation.
- Consul General. Fast memory manager based on record length.
- Mark and sweep.
- Reference accounting. A simple distribution manager that counts the number of relationships in data and extracts space when it is zero.
Distributed systems
- Lamport orders. Event handler based on the happened-before relationship (arrived earlier).
- Snapshot. This is a backup of the overall state of the system.
- Vector locks. Full schedule of events.
- Marzullo. Distributed timing synchronization.
- Intersection. Another algorithm based on allotted time.
Operating Systems
- Banker. Algorithm to avoid landings.
- Repeating page. Select pages to sacrifice when memory is missing.
- Bulli. Selection of priority position.
Disk management algorithms.
- Elevator. Planning the operation of the disk as an elevator.
- Short first. Schedule disk to reduce access time.
Process synchronization algorithms.
- Peterson. Allows two processes to share the same resource without conflicts, using shared memory for data exchange.
- Lamport Bakery. Improves the reliability of managing multiple multi-tasking processes with mutual exceptions.
- Dekker. Another competing programming algorithm, like Peterson's.
Mining algorithms (layout)
- Earliest deadline first planner. When an event occurs (end of task, new task, etc.), a process can be found in the list that should end as early as possible.
- Fair-share scheduling. Divides processor time between groups or users. To do this, another algorithm for managing sharing between processes is called recursively.
- Least slack time scheduling или Least Laxity First. Affects priorities based on time differences for processes. term, time of readiness, time of execution.
- List of schedulers. From the list of processes with priorities, first of all allocates available resources to the most priority areas. Possible strategies. critical path, longer path, first level, more processing time.
- Multilevel feedback.
- Rate-monotonic scheduling. Optimal, proactive algorithm with static priority. Priority is given on the principle of a monotonous rate (the first ends with the first contract).
- Round-Robin scheduling. Easiest - assigns time slices to each process without priority.
- Short work next (or first). Then executes the pending process, which has the shortest execution time, without preemption.
- Shorty repainted the time. The version of the study of the shortest upcoming process that completes the task in the process before choosing another.
Geometry
- Gift wrapping. Defines the convex hull of a set of points.
- Gilbert-Johnson-Curty distance. Find the shortest distance between two convex shapes.
- Graham scanned. Defines the convex hull of a set of points on a plane.
- Line segment intersection. Find out which lines intersect the imaginary line.
- Points in a polygon.
- Rey/plane intersection. Intersection of rays with a plane.
- Line/triangle intersection. Ray/Ploy Special Case.
- Implicit surface polygon. Approximates the surface implied by the polygon representation.
- Triangulation. A method of estimating distance or determining properties of a topological space .
Columns
- A * three searches. Find the best path between two nodes in the graph. Improved best-first, using heuristics to speed up searches.
- Bellman Ford. Calculates the shortest paths in a value graph (or arcs can have a negative weight).
- Graph canonization. It consists in finding the canonical form of a graph in such a way that it is isomorphic to another graph. Used in chemoinformatics.
- Dijkstra algorithm. Calculates the shortest paths in a graph without a negative arc.
- Violation of methods. Calculates the shortest paths in a graph.
- Floyd-Warshall. Solve the smallest path problem in the oriented and currency graph.
- Floyd cycle-search. Iterative loop-defining algorithm.
- Hopcroft-Karp algorithm. Returns the maximum of dimensions without common points from a two-part graph. Alternatives are algos breadth-first and depth-first.
- Johnson. Short path in oriented graphics.
- Kruskal. Find the minimum path tree for the graph.
- Approx. Robert Prim's algorithm finds the minimum path tree of the graph. Also called DJP, Yarnik or Prim-Yarnik.
- Borówka. Like Kruskal.
- Ford-Fulkerson. Calculates the maximum flow in a graph.
- Edmonds-Karp. Ford-Fulkerson implementation.
- Non blocking Minimal Spanning Switch. Telephone exchange.
- Woodhouse Sharp. Find the minimum path tree for the graph.
- Based on spring. Graph drawing algorithm.
- Hungarian. Find the perfect match.
- Coloring. Graph coloring.
- Nearest neighbor. Find your nearest neighbor.
- It comes out topological. Classify an acyclic directed graph so that each node precedes the nodes with which it is associated (depending on the value of the arcs).
- Tarjan's off-line least common ancestors algorithm. Calculates the closest common ancestors to pairs of nodes in the tree.
Graphics
- Brazilian line. Uses solution variables to display a string between two points.
- Coloring. The process of coloring a black and white image or video with multiple characters to mark colors. Examinations.
- Division. Under the original name "Depixelizing Pixel Art," this anti-aliasing algorithm converts an image into coarse pixels into a realistic image. (Johannes Kopf and Dani Lischinsky). Demonstration. Demo Archive. Implementation in C++.
- HDR. There are many photo contrast algorithms.
- DDA line algorithm. Floating point math is used to draw a line between two points.
- Flood phill. Paints an area bounded by a closed line.
- Xiaolin Wu line algorithm. Line smoothing algorithm.
- Painter's algorithm. Detects visible parts of the 3D scene.
- Ray traces. "Rendering," realistic drawing.
- Fong shading. Creates lighting in 3D.
- Guraud shading. Simulates the effect of a lyometer and color on the surface of a 3D object.
- Scanline rendering. Built an image by moving an imaginary line.
- Global enlightenment. Restores direct and reflected lighting to other objects.
- Interpolation. Built a new point on the enlarged image.
- Resynthesist. Deletes the object in the photo, restoring the background. Used by Photoshop and The Gimp. Resynthesis tutorial.
- Sloop-intercept algorithm. It is an implementation of the slop-intercept formula for drawing a line.
- Spline animation. Minimizes mistakes with the Runge phenomenon.
- 3D Surface Tracker Technology. How to add images to walls in a video based on hidden surfaces.
Artificial intelligence
- Alpha-beta. Alpha max plus beta min. The main Algo is used to find the best punch in the table game, in particular (failure, lady, reversi).
- Syndicate analysis. In fact, the combination of Bayes algorithms, maximum entropy and SVM (vector-supported machine).
- Ant colonies. A set of algorithms based on the behavior of ants that achieve optimization to solve a problem.
- KLA. Cortical learning algorithm. For robotic learning based on three properties, disparate distributed representations, temporal inference, online learning. NuPic project code in Numenta.
- DE (Differentiated Evolution). Solve the polynomial Chebyshev problem, which refers to electronic filters.
- Semi-controlled receipt of sarcastic sentences in an online product review.
An algorithm that recognizes saccharism or irony in a tweet or online document. Such an algorithm would also be crucial for programming humanoid robots.
Computer vision
- Epitome. Represents a smaller image or video.
- The number of objects in the image. Uses the connected component algorithm to identify each object and count the objects.
- Deep Dense Face Detector. Farafad, Saberian and Lee. Able to recognize faces from different angles .
- Evolution - structural elements. Brigham Young University. For the ability to identify known objects and add new objects to the knowledge base without human intervention.
- O'Carroll's algorithm. Based on a transformation into insect vision mathematics, this algorithm estimates how to move while avoiding objects.
- Learning tracking detection. Detects and tracks moving objects.
- Viola-Jones, a simple and fast object detection framework. It is capable of face recognition and is implemented in OpenCV.
Genetic algorithms
They use three operators: Selection (solution selection), reproduction (using the selected solution to build others), replacement (replacing the solution with the best).- Fitness is a proportional choice. Also called roulette, choose solutions.
- Select a node line. Selection of decisions classified by relevance.
- Turning selection. Choose the best solution through some tournament.
- Stochastic universal sampling. Individuals are associated with adjacent line elements so that each has a corresponding size.
Training
- PAVA (Pool-Contricent-Violators Algithm). Leu, Hornik, Mair. Improves the function for a set of points. Isotonic regression optimization. C++ code.
- Weight Multipliers. Assigns weight to different decision strategies. This refers to the recognition of forms and is found in natural genetics .
Neural networks
- Hopfield network. A repeating neural network operating as binary address memory. It converges into a stable state.
- Backpropagation. The training method helped train artificial neural networks.
- Kohonen map. Trained neural networks using autonomous learning to produce a low-profile representation (2D, 3D) of the examples given.
Lists, Tables, and Trees
Research
- The tree is upside down (Splay tree). A binary tree with a reversal function to place the node at the root and reorganize the rest accordingly.
- Dictionary search. See predictive search.
- Algorithm selection. Finds the largest level k entry in the list.
- Dichotomous search (binary search). Finds an object in an ordered list.
- Bredt-first search. Crosses level by level graph.
- Depth-first search. Crosses a graph with a branch.
- Best search. Traverses the graph in order of probable importance using a priority queue.
- Separate set or search-union (union-search). Using the creation of a labyrinth.
- Single-cheap search. A tree search that allows you to find the lowest cost if costs change.
- Predictive search. A kind of dichotomous research.
- Hash table. Associates keys with items to search in an unclassified list of linear duration.
- Interpolated search. See predictive studies.
- Skip list. A structure consisting of chain lists for quick access, and a search/insert algorithm.
- Median. Search for the median in the list of extraordinary numbers. Algo Torbena is less quick, but doesn't change the picture.
Classification
- A binary tree comes out. Binary tree ranking, incremental, similar to insertion sorting.
- Bogosort. Ineffective rating of a stack of cards by random selection.
- Sorting the bubble. For each pair of indices, elements that are out of order interact.
- Bucket comes out. Splits the list into batches and sorts them separately. Generalizes blue.
- A cocktail (or bi-directional bubble sorting) comes out. Sort in two directions at once.
- Combe comes out. Efficient bubble sorting that eliminates small values at the end of the sort and uses spaces between values.
- Countdown is coming out. Uses intervals between numbers in list A to create list B. Indexes B are used to determine A values less than i.
- Gnome comes out. As sorting by insertion, but moving to the right place occurs using a series of inversions, as in bubble sorting.
- Heapsort. Converts the list to a stack, removes the largest item, and appends it to the end of the list.
- Tim comes out. Analyzes the list to be classified before selecting the optimal procedure. Probably the fastest and does not depend on the size of the list. Used inside Java, Python, C++.
- Sort by Insert. Locates the current item in the classified list and places it in the classified list.
- Introsort or Introspective is coming out. Starts at quicksort and switches to heapsort when a certain level of recursion is reached.
- Murge comes out. Classifies the first and second halves of the list separately and combines them (recursively).
- Pancake comes out. Interacts with elements of any prefix in the sequence.
- Dove comes out. Populates an empty table with the items you want to classify in another table, in order.
- Postman comes out. Changing the sorting of packages, multi-level, is used by post offices for mail.
- Quicksort. Divides the list into two parts, with all elements of the first lower than the second, and classes (recursively).
- Sort by bases or sort radis. (Radix is coming out). Classifies keys associated with list items or integers from digits.
- The choice comes out. Takes the last remaining item and adds it to the end of the classified list.
- Shell comes out. Improved insertion sorting using value spacing.
- Smutsort. See hipsort.
- Stochastic comes out. Look, bogosort.
Merge
- Just Murge. Combines n sorted streams into one. The flows are compared, the smallest inputs of each are sent to the outgoing flow.
- Tri k-way Merge (or p-way). Sorts the data stream by repeating merges.
Mathematics
Algebra
- Buchberger algorithm. Find Greibner's base.
- Eigeneum algorithm.
- Advanced Euclidean algorithm. Solve the equation ax + by = c.
- Fourier multiplication. For very large numbers, calculate the fast Fourier transform of two numbers and multiply both results by.
- Gram-Schmidt process. Orthogonalizes a set of vectors.
- Gauss-Jordan was eliminated. Solve the system of linear equations.
- Karatsuba multiplication. Efficient recursive algorithm for large numbers. Toom-Cook derivative.
- Knut-Bendix complements. To rewrite the system of rules.
- Multivariate Department. Polynomials over several uncertainties.
- Risch algorithm. Translates an indefinite integral into an algebraic problem.
- Toom-Cook (Toom3). Splits each number into several parts.
Eigenalue algorithm
"Eigenvalue" and/or "Eigenvector" matrix search algorithms.
- QR algorithm. A popular method based on QR decomposition.
- Reverse iteration. Iterative algorithm.
- Reilly iteration factor. Propagates the inverse iteration principle by using the Reilly coefficient to gradually obtain sufficient eigenate estimates.
- Arnoldi iteration. Computes eigenetic estimates of orthogonal projection A in Krylov subspace.
- Lanczos iteration. Method for finding a zero verzer during quadratic filtering.
- Jacobi method. Digital procedure for calculating eigenerators and eigenvectors of a real symmetric matrix.
- Bisection.
- Divide-and-end. Used for real symmetric matrices.
eigenvector algorithms
- Richardson's eigenvector algorithm.
- Max-Plus eigenvector algorithm. For non-linear H1 control.
- Abrams and Lloyd eigenvector algorithm.
Arithmetics
- Binary PGCD. Calculate the largest common divisor quickly.
- Booth multiplication. Multiplies two integers using the complement of 2.
- Euclidean algorithm. Calculates the PGCD.
- Binary (or Egyptian) multiplication. The largest multiple is divided by the sum of the power equal to two, and a table of the forces of two of the second is created.
Latent logarithm in group theory
- Baby-step giant-step. These are a number of known steps for calculating the logarithm.
- Pollard's rho algorithm for logarithms. An analogue of the algorithm of the same name for integer factorization, which can also be applied to the calculation of a single logarithm.
- Pohlig-Hellman algorithm. Solving the problem for a multiplicative group whose order is an integer. Based on the Chinese remainder theorem, and holds in polynomial duration.
- Calculated algorithm index. A well-known algorithm for some groups, such as the multiplicative group modulo m.
All factorization
Divide the integer into primary factors.
- Fermat factorization method. Representing an even integer as the difference of two squares.
- Trial division. The simplest algorithm. trying to divide an integer into consecutive primes.
- Lenstra elliptic curve factorization или elliptic curve factorization method (ECM). Fast, requiring sub-exponential duration of elliptic curves.
- Pollard rho. Changing Pollard p-1 is effective for breaking numbers into a small number of factors.
- Pollard p-1. A split algorithm that is only suitable for integers with certain types of factors.
- Congruence of squares. Finding the congruence of modulo n squares is a way of factorizing the integer n.
- Square sive. Use the idea of Dixon's method. This is an algorithm that is simpler than the "number field," and the fastest for integers less than 100 digits.
- Dixon. Common use factorization method.
- Special field number. An ideal specialized algorithm for factorizing Fermat numbers.
- General sieve field number (GNS). Derived from "special field number." Effective for large numbers. Step by step.
First number test
Determine if the given number is prime.
- AKS primality test (Agrawal-Kayal-Saxena). The first published algorithm that is both polynomial, deterministic and unconditional. A generalization of Fermat's theorem extended to polynomials.
- Fermat primality test. Set to True Equality or Set of Equals for primary values to then check if they contain or name the number being tested.
- Miller-Rabin primality test. Sounds like Fermat's test. Inappropriate and probabilistic algorithm. About
- Eratoshene. An ancient algorithm for finding primes up to a certain integer.
- Sive of Atkin. Optimized version of Eratosten criterion.
- Nightingale-Strassen primary test. Same principle as Fermat's test.
Digital form
- Fibonacci. Fiboslavci sequence computation.
- Arc-gradient method. Linear equation resolution.
- Lynx dances. Find all solutions to the problem of accurate coverage.
- From Boer's algorithm. Curve calculations.
- От Casteljau's algorithm. Calculates Bezier curves.
- False method position. Approximation of function roots.
- Gauss-Legendre. Calculates PI decimal places.
- Kahan conceit. Efficient floating point addition.
- BET. Digital integration, Monte Carlo imitation.
- Newton method. Search for null function values.
- Rounds. Classical rounding of integers.
- Dry method. Approximation of function roots.
- Shifting nth-root. Displays the root by numbers.
- Root Square. Approximation of the square root of a number.
- Borwein algorithm. Evaluates to 1/e.
Statistics
- Metropolis-Hastings. Creates a sequence of patterns for a probable distribution of a variable or more.
Bases
Matrix calculation or optimization .- Squaring exponentation. Calculates the power of matrices.
- Rutishauser. Tridiagonization of matrices.
- Strassen algorithm. Fast matrix multiplication.
- Symbolic Cholesky decomposition. Efficient storage of matrices.
- Algorithm Zh. Improves Rutishauser for tridiagonizing oriented matrices.
- Multiplication array string. Optimizing matrix sequence multiplication in the most efficient way through dynamic programming.
Optics
- Gerchberg Saxton. Determination of phase from image and diffraction plane.
Optimization and online search
See also Graphs.
- Optimal organization. Place the largest number of objects on a bounded surface.
- Ant colonization optimization. Probabilistic technique for solving problems that boil down to finding the right path on a graph.
- BFGS (Broyden-Fletcher-Goldfarb-Shanno method). Solving the nonlinear optimization problem without constraints.
- Bellman Ford. (1955). Calculates the shortest path between a vertex and all other vertices on a directed graph. Supports negative values, unlike Dijkstra's algo.
- Brunch and bound. (Separation and evaluation.) A method for finding the optimal solution to problems of restrained and combinatorial optimization.
- Combined gradient method. An iterative algorithm for digitally solving systems of linear equations whose matrix is symmetric and positive.
- Strategic evolution. An optimization technique based on the idea of adaptation and evolution. The operators are: selection, recombination, mutation, environmental selection.
- Maximum flow Almost linear. Algorithm by Kelner, Tat Lee, Orecchia, Sidford to get maximum flow by considering all paths simultaneously.
- Gauss-Newton. Solve the problem of the smallest nonlinear square.
- Downgrade. Approaches the local minimum of a function in steps proportional to the negative degree (or approximation) of the function at the current point.
- Gradient up. An approach to the local maximum of a function, like a "gradient," with steps directly proportional to the gradient.
- Hungarian (Kun-Munkre). Optimizes resource or workstation allocation at lower cost.
- Levenberg-Markardt. Like Gauss-Newton.
- Linear search. Iterative approach to find the local minimum of an element when optimizing without constraints.
- Local search. Meta-heuristics for solving complex optimization problems such as criterion maximization among multiple candidate solutions.
- The problem of stable marriages. It is used in economics, mathematics, biology and other sciences. We want to link the set x of incompatible elements and the set y of incompatible elements so that each x finds the best y for it and vice versa. JavaScript source.
- Nelder-Mead method (downhill simplex). Nonlinear optimization.
- Newton's method in optimization. The same algorithm that finds the roots of equations in one or more dimensions can also be used to find local minima and maxima of functions.
- Paxos. A set of distributed algorithms to build consensus among multiple sentences and multiple factors.
- PSO, Particle Swarm Optimization. Intelligence of a swarm modeled by particles in multidimensional space, with location and speed.
- Random-restart hill climbing. Meta algorithm built on the "hill climbing optimization" algorithm.
- Simplex. Solving the linear programming problem.
- Simulated year. The generalized and probabilistic meta-algorithm for the global optimization problem is based on the principle of annealing in metallurgy.
- Stipest is behind. See demotion.
- Stochastic tunnel. A function minimization approach based on the Monte Carlo sampling method.
- Tabu search. Optimize search with storage structures.
- Trust search. Another iterative approach to finding the local minimum of an element when optimizing without constraints.
Parsing
- CEK algorithm. An efficient O (n3) algorithm for any non-contact grammar.
- Earley's algorithm. Another O (n3) algorithm for non-contact grams.
- Inside-out. The O (n3) algorithm overestimates the probability of staging in non-contact probabilistic grammars.
LL Parsers
Parses non-contact grammar from top to bottom, from left to right. Like ANTLR, which is LL (*).LR Parsers
Parse non-contact grammar from bottom to top, so starting with the last descendants of each branch .- Dijkstra's shunting yard algorithm. Usually used to implement parsers preceding the operator, which convert incorrect notation into reverse Polish notation.
- LALR (Look-Ahead LR). They go from left to right with the reading of one token. Jakk and Beeson are LALR parsers (1). The Social Revolutionaries are superior.
- SLR (simple LR) parser. LR (0) analyzer modified to avoid offset and decrement collisions. Remains below LR (1).
- Canonical LR parser or LR (1). You will see the token in advance.
- HLR. (Generalized LR parser) by Masaru Tomita. LR extension to manage nondeterministic or ambiguous grammars. It is effective for natural language.
Recursive Parsers Down
LL parsers are constructed from a set of mutually exclusive procedures representing the rules of grammar production.- Pakrat. Linear duration algorithm for non-contact grams LL (k). Copies and remembers choices to avoid missing them.
Forecasting (statistics)
- Baum-Welch. Finds unknown parameters of the hidden Markov model (HMM) using the back-and-forth algorithm.
- Viterbi. Computes the Viterbi path, the sequence of states that has the highest probability of appearing in a sequence of events.
Logic programming
- Davis-Putnam algorithm. Check the validity of the first-order statement.
Part
Applying quantum calculations to various problems
- Grover's algorithm. Provides quadratic acceleration for multiple search problems.
- Shore's algorithm. Provides exponential acceleration when calculating a number.
- Deutsch-Jozsa. Balance criterion for Boolean function.
Sciences
Astronomy
- Ephemeris.
- Moon positions and other celestial objects .
- Julian Day. The number of days since Monday, January 1, 4713 BC in the proleptic Julian calendar. An algorithm is a formula. We have options. heliocentric, chronological, modified, abbreviated, truncated Julian day, Dublin, Lilian.
- Julian date. Julian day is non-random, with decimal fraction.
Medical
- Calculations for drugs .
- Diagnostic assistance.
(Processing) Signal
- CORDIC. Fast trigonometric function.
- Rainbow flow. Reduces the history of complex strands to a count of inversions of elementary strands.
- Osem. Medical image processing.
- Herzel algorithm. Can be used to decrypt DTMF.
- Cautious Fourier. Identifies the frequencies in the signal segment .
- Fast Fourier transform
- Cooley-Tukey FFT
- Rader's FFT
- Bluestein's FFT
- Bruun's FFT
- FFT prime factor
- Richardson-Lucy deconvolution. Image selection algorithm.
- Elser Difference-Map. Microscopic X-ray diffraction.
- Shazam. Recognizing a piece of music by comparing the signal and determining its specifics.
Text
Research
- Aho-Corasik. Search the text by building a table of words.
- Bitap (or shift-or, shift-and, Baeza-Yates-Gonnet). Fuzzy search algorithm developed by Udi Manber and Sun Wu.
- Boyer-Moore string search. Search the text by skipping substrings that do not contain the letters of the search word .
- Burrows Wheeler transforms. A string transformation that can be used to find words in text more quickly.
- Knut-Morris-Pratt. Creates a table during a search to skip substrings.
- The problem of the longest common subsystem. (Haskell algorithm). It has two sequences.
- The longer subsequence increases. Common to two sequences. This boils down to finding the longest path in a graph oriented without cycles.
- Shortest overall super sequence. It has two sequences.
- Rabin-Karp string search. The hash is used for several searches.
- Horspool. Boyer-Moore algorithm simplification. O (min).
Comparison with approximation
- Distance from Lowenstein (or edit distance). Minimum number of operations (insert, delete, replace) to convert a word to another word.
- Soundex. Phonetic algorithm for indexing words by their sound (in English).
- Metaphon. Indexing words by sound (in English).
- NYSIS. (New York State Identification and Intelligence System). Phonetic algorithm that improves soundex.
Word processing
- Latent Provisioning (LDA). Analyze documents to associate content with context. Used by search engines.
- Latent semantic indexing (LSI). Or hidden semantic indexing. Automate methods of attaching text to a context from words that typically appear in that context.
- Stemming. A method of reducing words to their stems or roots.
Utilities
- Doomsday. Day of the week.
- Xor exchange. Changes the values of two variables without an intermediate variable.
- Humming in weight. Finds the number of bits 1 in a binary word.
- Moon. Method of checking identification numbers.
- Create bit mask. Bit manipulation algorithm.
Misc
- BrowseRank. An alternative to PageRank based on user behavior.
- Leaf shape. Of the 28 parameters, this algorithm reconstructs the shape of all leaves produced by nature.
- Hypertext Induced Topic Selection (HITS, 1997 patent). Algorithm for calculating web page scores by Jon Kleinberg. One score depends on the incoming and the other on the outgoing.
- PageRank. (1998) Algorithm for evaluating pages by Larry Page and Sergey Brin (Google) from incoming and outgoing links. The score also depends on many other criteria, see Google's patent for evaluating pages.
- Schreier-Sims. Permutation of groups. BSGS (Base and Strong Generating Set) calculation method. Used by algebraic algorithms.
- Robinson-Shenstead. Combinatorial algorithm.
Communications
- Image conversion algorithms,
Last updated: March 16, 2021.
Legal: You are free to print this page. Do not use it on another site, but post a link to this page.