R Language Machine Learning package

Source: Internet
Author: User
Tags svm

From:http://www.zhizhihu.com/html/y2009/410.html

Machine learning is an interdisciplinary area of computer science and statistics, and R on machine learning consists of the following aspects:
1) Neural Network (neural Networks):
The Nnet packet performs a single hidden layer feedforward neural network, and Nnet is part of the VR package (http://cran.r-project.org/web/packages/VR/index.html).
2) Recursive splitting (Recursive partitioning):
Recursive splitting uses the tree structure model to do regression, classification and survival analysis, mainly in Rpart packet (http://cran.r-project.org/web/packages/rpart/index.html) and Tree packet (HTTP/ cran.r-project.org/web/packages/tree/index.html), especially recommended Rpart packages. Weka also has this recursive splitting method, such as: J4.8, C4.5, M5, packet Rweka provides an interface for functions of R and Weka (http://cran.r-project.org/web/packages/RWeka/index.html).
The party package provides a two-class recursive split algorithm that can be used for unbiased variable selection and stop criteria: the function ctree () detects the relationship between the independent variable and the dependent variable by the non-parametric conditional inference method, and the function mob () is able to establish a parametric model (http://cran.r-project.org/web /packages/party/index.html). In addition, the party package also provides a visual display of two branch tree and node distribution.
The Mvpart package is a Rpart improvement package that handles the problem of multivariate dependent variables (http://cran.r-project.org/web/packages/mvpart/index.html). Rpart.permutation Packet Replacement Method (permutation) evaluation Tree Effectiveness (http://cran.r-project.org/web/packages/rpart.permutation/ index.html). Knntree package establishes a classification tree, each leaf node is a KNN classifier (http://cran.r-project.org/web/packages/knnTree/index.html). The Logicreg package does logistic regression analysis for most of the arguments are two variables (http://cran.r-project.org/web/packages/LogicReg/index.html). Maptree Pack (http://cran.r-project.org/web/packages/maptree/index.html) and Pinktoe pack (http://cran.r-project.org/web/ packages/pinktoe/index.html) provides a visual function of the tree structure.
3) Stochastic forest (random forests):
The Randomforest package provides functions for regression and classification using random forests (http://cran.r-project.org/web/packages/randomForest/index.html). Ipred pack bagging thought to do regression, classification and survival analysis, combining multiple models (http://cran.r-project.org/web/packages/ipred/index.html). The party package also provides a random forest law (http://cran.r-project.org/web/packages/party/index.html) based on the conditional inference tree. VARSELRF Pack Random forest Law to do variable selection (http://cran.r-project.org/web/packages/varSelRF/index.html).
4) regularized and Shrinkage Methods:
LASSO2 Pack (http://cran.r-project.org/web/packages/lasso2/index.html) and Lars Pack (http://cran.r-project.org/web/packages /lars/index.html) can execute a regression model in which the parameters are subject to certain limitations. The Elasticnet package calculates all shrinkage parameters (http://cran.r-project.org/web/packages/elasticnet/index.html). Glmpath package can get generalized linear model and Cox model of L1 regularization path (http://cran.r-project.org/web/packages/glmpath/index.html). Penalized package Execution Lasso (L1) and Ridge (L2) penalty regression model (penalized regression models) (http://cran.r-project.org/web/packages/ penalized/index.html). The PAMR package performs a reduced centroid taxonomy (shrunken centroids classifier) (http://cran.r-project.org/web/packages/pamr/index.html). The earth package can do multiple adaptive spline regression (multivariate adaptive regression splines) (http://cran.r-project.org/web/packages/earth/ index.html).
5) Boosting:
GBM Pack (http://cran.r-project.org/web/packages/gbm/index.html) and boost pack (http://cran.r-project.org/web/packages/ boost/index.html) performs a variety of gradient boosting algorithms, GBM packages do tree-based gradient descent boosting,boost packets including Logitboost and L2boost. The Gammoost package provides programs based on the boosting generalized additive model (generalized additive models) (http://cran.r-project.org/web/packages/GAMMoost/ index.html). The Mboost package makes model-based boosting (http://cran.r-project.org/web/packages/mboost/index.html).
6) Support vector machines (supported vectors machines):
The SVM () function of the e1071 package provides an interface for R and LIBSVM (http://cran.r-project.org/web/packages/e1071/index.html). The Kernlab package provides a flexible framework for learning methods based on kernel functions, including SVM, RVM ... (http://cran.r-project.org/web/packages/kernlab/index.html). The Klar package provides an interface for R and Svmlight (http://cran.r-project.org/web/packages/klaR/index.html).
7) Bayesian method (Bayesian Methods):
Bayestree package execution Bayesian Additive Regression Trees (BART) algorithm (http://cran.r-project.org/web/packages/BayesTree/ Index.html,http://www-stat.wharton.upenn.edu/~edgeorge/research_papers/bart%206--06.pdf). TGP Bayesian Semi-parametric nonlinear regression (Bayesian nonstationary, semiparametric nonlinear regression) (http://cran.r-project.org/web/ packages/tgp/index.html).
8) Optimization based on genetic algorithm (optimization using genetic algorithms):
Gafit Pack (http://cran.r-project.org/web/packages/gafit/index.html) and Rgenoud pack (http://cran.r-project.org/web/ packages/rgenoud/index.html) provides optimization programs based on genetic algorithms.
9) Association Rule (Association Rules):
The Arules package provides a data structure that effectively handles sparse two metadata, and provides function Apriori and Eclat algorithms for mining frequent itemsets, maximum frequent itemsets, closed frequent itemsets, and association rules (http://cran.r-project.org/web/ packages/arules/index.html).
10) Model selection and validation (models selection and validation):
The tune () function of the e1071 package selects the appropriate parameter (http://cran.r-project.org/web/packages/e1071/index.html) within the specified range. The Errorest () function of the ipred package uses a resampling method (cross-validation, bootstrap) to estimate the classification error rate (http://cran.r-project.org/web/packages/ipred/index.html). The function in the Svmpath package can be used to select the cost parameter C (http://cran.r-project.org/web/packages/svmpath/index.html) of the support vector machine. The ROCR package provides functions for visualizing the performance of the classifier, such as the ROC Curve (http://cran.r-project.org/web/packages/ROCR/index.html). The caret package provides a variety of functions for establishing predictive models, including parameter selection and importance measurement (http://cran.r-project.org/web/packages/caret/index.html). CARETLSF Pack (http://cran.r-project.org/web/packages/caretLSF/index.html) and CARETNWS (http://cran.r-project.org/web/ packages/caretnws/index.html) package provides functionality similar to the caret package.
11) Basis of statistical learning (Elements of statistical learning):
Book "The Elements of statistical learning:data Mining, inference, and prediction" (http://www-stat.stanford.edu/~tibs/ elemstatlearn/) data sets, functions, and examples are packaged in the Elemstatlearn package (http://cran.r-project.org/web/packages/ElemStatLearn/ index.html).

Website: http://cran.r-project.org/web/views/MachineLearning.html Maintenance staff: Torsten Hothorn

R Language Machine Learning package

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.