Downloaded 3,948 times
![Box plot >par(mfrow=c(1,2)) > plot(iris$Petal.Length) >boxplot(iris$Petal.Length~ iris$Species) > par(mfrow=c(2,2)) # to draw four figs in one window > for(i in 1:4) boxplot(iris[,i] ~ Species, data=iris, main=names(iris)[i])](https://image.slidesharecdn.com/irisdataanalysiswithr-140801203600-phpapp02/75/Iris-data-analysis-example-in-R-13-2048.jpg)
![Histogram > par(mfrow=c(1,1)) >hist(iris$Petal.Length[1:50]) Subsetting: >iris$Sepal.Length[1:50] >iris$Sepal.Length[-(1:50)] Select by name: >iris$Sepal.Length[iris$Species == "setosa"] Change the order of data frame: >iris.ordered<-iris[order(iris$Sepal.Length),]](https://image.slidesharecdn.com/irisdataanalysiswithr-140801203600-phpapp02/75/Iris-data-analysis-example-in-R-18-2048.jpg)
Uploaded byDuyen Do
Iris data analysis example in R
This document discusses analyzing the Iris flower data set using R. It provides an overview of the Iris data, which contains measurements of Iris flowers from three species. Various data exploration techniques are demonstrated, including scatter plots, box plots, histograms and outlier detection. Clustering, classification and regression algorithms are explored, such as k-means clustering, Fisher's linear discriminant analysis, and linear regression. The document serves as a tutorial for analyzing a sample data set using common statistical and machine learning methods in R.
Iris data analysis example in R
- 1.
- 2. Overview: data analysisprocess
- 3. Iris setosa Iris virginica Irisversicolor
- 4. Iris flower dataset • Also called Fisher’s Iris data set or Anderson’s Iris data set • Collected by Edgar Anderson and Gaspé Peninsula • To quantify the morphologic variation of Iris flowers of three related species • >iris
- 5. Draw a hypothesisthat you can test! • Null hypothesis • Alternative hypothesis • P-value < 0.05
- 6. Get data! Some waysto read data in R: • read.table, read.csv, read.xls, data.frame,.. • edit,… • … => Hint: Never modify your raw data file; always work on a copy!
- 7.
- 8. Some basic functionin R to examine iris data: >?iris >names(iris) >iris >str(iris) >iris$new_class_specis<-as.character(iris$Species) >iris$new_class_specis<-NULL >iris$Species <- gsub("%","",iris$Species)) >iris<-na.omit(iris)
- 9. Summarize and plotyour data! >summary(iris) >plot(iris)
- 10.
- 11.
- 12.
- 13. Box plot >par(mfrow=c(1,2)) > plot(iris$Petal.Length) >boxplot(iris$Petal.Length~iris$Species) > par(mfrow=c(2,2)) # to draw four figs in one window > for(i in 1:4) boxplot(iris[,i] ~ Species, data=iris, main=names(iris)[i])
- 14.
- 15. Histogram vs Barchart
- 16.
- 17.
- 18. Histogram > par(mfrow=c(1,1)) >hist(iris$Petal.Length[1:50]) Subsetting: >iris$Sepal.Length[1:50] >iris$Sepal.Length[-(1:50)] Select byname: >iris$Sepal.Length[iris$Species == "setosa"] Change the order of data frame: >iris.ordered<-iris[order(iris$Sepal.Length),]
- 19. Build a statisticalmodel! Data mining: • Predict: • Classification • Regression • Deviation detection • Descript: • Clustering • Association Rule Discovery
- 20.
- 21.
- 22. Clustering • Principle: basedon measure of distances • Algorithms: • Hierarchical clustering: bottom up, top down • Centroid-based clustering: k-mean, PAM, CLARA, CLARANS,.. • Distribution-based clustering: STING, WAVECluster, CLIQUE,.. • Density-based clustering: DBSCANS, OPTICS, DENCLUE,.. • Model-based cluatering: statistical model + Neural network • …
- 23.
- 24.
- 25. Clasification algorithms • Linearclassifiers: Fisher's linear discriminant analysis, Naive Bayes classifier,.. • Support vector machines: Least squares support vector • Quadratic classifiers • Kernel estimation: k-nearest neighbor • Boosting (meta-algorithm) • Decision trees: Random forests • …
- 26. Fisher's linear discriminantanalysis (LDA) Demo in R
- 27. Regression analysis Y ≈f(X, β)
- 28. Regression analysis • Methods:Linear regression, Logistic regression, Poisson regression • Regression analysis is widely used for prediction and forecasting
- 29.
- 30.
- 31. Analysis simple linearregression in R Demo in R
- 32. Estimate • Clustering: hard,using: user examine, similarity measure, classification algorithms, entropy, F-measure, pure,… • Clasification: holdout, k-fold cross validation,… • Regression: statistical hypothesis test
- 33.
- 34. References • http://en.wikipedia.org/wiki/Iris_flower_data_set • http://ykhoa.net/r/R/Chuong%2010.%20%20Phan%20tich% 20hoi%20qui%20tuyen%20tinh.pdf •http://www.statsoft.com/Textbook/Elementary-Statistics- Concepts • http://bis.net.vn/forums/p/366/628.aspx
- 35.
- 36.