ALTHEA

Loading

ALTHEA

Category 存档

存档

PubMed

Explore PubMed is a free search engine that provides access to more than 36 million citations for biomedical literature from MEDLINE, life science journals, and online books.

It is maintained by the United States National Library of Medicine (NLM) at the National Institutes of Health as part of the Entrez system of information retrieval.

PubMed is a platform for scholars, researchers, and students of all kinds to discover, share, and connect valuable ideas. It is a tool to search MEDLINE, which is the database of biomedical research articles compiled by the U.S. National Library of Medicine, National Institutes of Health.

MEDLINE has over 21 million citations, with more being added every day.

存档

JSTOR

JSTOR is a digital library that provides access to more than 12 million academic journal articles, books, images, shared collections, and primary sources in 75 disciplines 12.

It is a platform for scholars, researchers, and students of all kinds to discover, share, and connect valuable ideas.

JSTOR is part of ITHAKA, a not-for-profit organization that also includes Artstor, Ithaka S+R, and Portico 1.

JSTOR offers free and low-cost access to scholarly content for libraries, researchers, and publishers 1.

存档

K-means vs SOM (3)

Iterative process:

K-means: Updates the cluster centers by iteratively calculating the distance from data points to the nearest cluster center.

SOM: Iteratively adjusts the weights of neurons through competitive learning and neighborhood adjustment.

Sensitivity to initial values:

K-means: Sensitive to the choice of initial clustering centers, different initial values may lead to different results.

SOM: Relatively less sensitive to the choice of initial weights, but the initial network topology needs to be chosen carefully.

Interpretability:

K-means: results are more intuitive, with each cluster represented by a center point.

SOM: Provides a topology of the data that can be used to visualize and understand high dimensional data.[label][/label][checkbox checked=”true/false”][/checkbox]

存档

K-means vs SOM (2)

Topology:

K-means: K-means does not consider topological relationships between data points and focuses only on the location of cluster centers.

SOM: SOM considers the topological relationship between data points on a grid of neurons, preserving the spatial information between neighboring neurons.

Applicable scenarios:

K-means: Suitable for spherical, equal-sized clusters with similar density.

SOM: Suitable for capturing topological structures and nonlinear relationships in data, especially for visualization of high-dimensional data.

存档

NSGA-Ⅱ

Step1: Initialize the population, randomly generate a set of individuals that meet the fitness requirements as the initial population.

Step2: Non-dominated sorting, according to the objective function value of individuals, the population is non-dominated. The non-dominated ordering divides the individuals into different ranks, and the individuals in each rank are not dominated by the more superior individuals.

Step3: Crowding Calculation, In order to maintain the diversity of the population, NSGA-II uses the crowding distance to measure the distribution density of individuals in the target space. Crowding degree distance indicates the density of target vectors around an individual, and a higher crowding degree distance indicates a more dispersed individual.

Step4: Selection operation to select a certain number of individuals as parents based on non-dominated ordering and crowding degree distance.

Step5: Crossover and mutation, perform crossover and mutation operations on the selected parent individuals to generate new offspring individuals.

Step6: Update the population, merge the parent and child individuals, and perform non-dominated sorting and crowding degree calculation according to the population size.

Repeat Steps 2-6 until the specified stopping condition is reached (e.g., number of iterations or optimal solution reached).

存档

K-means vs SOM

How it works:

K-means: K-means is an iterative, distance-based clustering algorithm. It iterates until convergence by assigning data points to the cluster centers closest to them and updating the cluster centers.

SOM: SOM is an unsupervised learning algorithm that works based on competitive learning. In SOM, a neuron grid is defined and the data points are mapped onto this grid and clusters are formed by competitive learning and adjustment of neighborhood relations.

Model Structure:

K-means: K-means divides the data points into K clusters, each represented by a cluster center.

SOM: SOM uses a two-dimensional lattice of neurons, where each neuron represents a cluster and neighboring neurons are similar in the feature space.

[label][/label][checkbox checked=”true/false”][/checkbox]