Please enable JavaScript.
Coggle requires JavaScript to display documents.
Main machine learning tasks and associated methods, Made by KAJJI Anass…
Main machine learning tasks and associated methods
Regression and classification
is a supervised ML=> to predict predifined class from a given set of attributes
Gaussian process, linear regression, K-nearest neighbors, ANN,SVM, Random Forests,decision trees,logistic regression, naive Bayes and deep learning
Clustering
is an unsupervised ML
=> partitions the input data set into subsets (clusters)
hierarchical clustering, partitioning methods
K-means, K-medoids, Mean-Shift
grid based clustering
CLIQUE, Sting, Wave Cluster
model based clustering
EM, COBWEB
density based methods
DBSCAN, Optics, Denclue
Association rules
find frequent patterns, correlations, associations or causal structures
Apriori, FP-growth and Eclat
Feature selection
reducing the number of attributes. Build simpler models ofbetter accuracy and can reduce overfitting
filter methods
LDA ,ANOVA, chi-square tests
wrapper methods
forward selection, backward selection, recursive feature elimination
embedded methods
mRMR, Greedy
Reinforcement learning
Learn optimal actions from a finite set ofavailable actions through continuously interacting with an unknown environment
Model-free
The optimal control policy is learned without first learning an explicit model
policy search
metaheuristics, policy gradient
value-function
Emporal fifference (TD) learning,
ACM, Q-learning, SARSA, DQN,DDPG
Model-based
uses a reduced number of interactions with the real environment during the learning phase.
Made by KAJJI Anass based on literature reviews and especially the Scientific paper intitled "Machine learning into metaheuristics: A survey and taxonomy of data-driven metaheuristics"