R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK...
-
Upload
shyanne-gibble -
Category
Documents
-
view
224 -
download
7
Transcript of R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK...
![Page 1: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/1.jpg)
R for Classification
Jennifer BroughtonShimadzu Research LaboratoryManchester, UK
2nd May 2013
![Page 2: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/2.jpg)
Classification?
Automatic Identification of Type (Class) of Object from Measured Variables (Features)
Object Type Feature1 Feature2 Feature3 …….Feature nLabel 1 val[1,1] val[1,2] val[1,3] ……. val[1,n]Label 2 val[2,1] val[2,2] val[2,3] …….val[2,n]…… ……. ……. ……. …….………Label m val[m,1] val[m.2]val[m,3] ……. val[m,n]
2 of 17
![Page 3: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/3.jpg)
Example Data
3 of 17
![Page 4: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/4.jpg)
Data Preparation & Investigation
EDA Technique
Box Plots PCA Decision Trees Clustering
Training Set
• Best features to distinguish between classes
• Relationships between features• Feature reduction
4 of 17
![Page 5: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/5.jpg)
Box Plots
PCA & Multivariate Analysis: ade4FactoMineR
5 of 17
![Page 6: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/6.jpg)
Example Classifier
6 of 17
![Page 7: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/7.jpg)
Classification Algorithms in R
Rattle: R Analytical Tool to Learn Easily (Rattle: A Data Mining GUI for R, Graham J Williams, The R Journal, 1(2):45-55 )7 of 17
![Page 8: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/8.jpg)
SVM
8 of 17
![Page 9: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/9.jpg)
Ensemble Algorithm
9 of 17
![Page 10: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/10.jpg)
Training and Testing
Classification Algorithm:
Neural NetworkSupport Vector MachineRandom Forest
Training Set(labelled)
Test Set(unlabelled)
TrainedClassifier
Classification Results
PredictionResults
+ Labels
Assess Predictions:Confusion MatrixROC Curve (2 categories) ….
10 of 17
![Page 11: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/11.jpg)
Using Classifiers in R
Select Training Data
Build Classifier
Run Classifier
classifier algorithm(formula, data, options)
(boosting and nnet)
classifier.pred predict(classifier, newdata, options)
11 of 17
![Page 12: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/12.jpg)
SVM & Neural Net Tuning
12 of 17
![Page 13: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/13.jpg)
Classifier Feedbackprint(classifier)plot(classifier)
high Gini Coefficient = high dispersion
13 of 17
![Page 14: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/14.jpg)
Classifier Prediction Resultspredict(type = “class”)
predict(type = “prob”)
confusion matrix14 of 17
![Page 15: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/15.jpg)
FalseNegative
TruePositive
TrueNegative
FalsePositive
Binary Classification Results
Y NClass Present?
ClassDetected?
Y
N
𝑻𝑷𝑹=𝑻𝑷
𝑻𝑷+𝑭𝑵=𝑺𝒆𝒏𝒔𝒊𝒕𝒊𝒗𝒊𝒕𝒚
𝑭𝑷𝑹=𝑭𝑷
𝑻𝑵+𝑭𝑷=𝟏−𝑺𝒑𝒆𝒄𝒊𝒇𝒊𝒄𝒊𝒕𝒚
15 of 17
![Page 16: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/16.jpg)
ROC Curves in RROCR package
16 of 17
![Page 17: R for Classification Jennifer Broughton Shimadzu Research Laboratory Manchester, UK jennifer.broughton@srlab.co.uk 2 nd May 2013.](https://reader035.fdocuments.us/reader035/viewer/2022062404/551b7743550346d31b8b63a6/html5/thumbnails/17.jpg)
Example Results
17 of 17