Лаба2. Анализ данных
Ангелина
2026-08-23
Задание 1. Пакет caret: getModelInfo() и featurePlot()
Установим caret, посмотрим какие методы в нём есть, и построим графики для тестовых данных.
library(caret)## Загрузка требуемого пакета: ggplot2## Загрузка требуемого пакета: latticenames(getModelInfo())## [1] "ada" "AdaBag" "AdaBoost.M1"
## [4] "adaboost" "amdai" "ANFIS"
## [7] "avNNet" "awnb" "awtan"
## [10] "bag" "bagEarth" "bagEarthGCV"
## [13] "bagFDA" "bagFDAGCV" "bam"
## [16] "bartMachine" "bayesglm" "binda"
## [19] "blackboost" "blasso" "blassoAveraged"
## [22] "bridge" "brnn" "BstLm"
## [25] "bstSm" "bstTree" "C5.0"
## [28] "C5.0Cost" "C5.0Rules" "C5.0Tree"
## [31] "cforest" "chaid" "CSimca"
## [34] "ctree" "ctree2" "cubist"
## [37] "dda" "deepboost" "DENFIS"
## [40] "dnn" "dwdLinear" "dwdPoly"
## [43] "dwdRadial" "earth" "elm"
## [46] "enet" "evtree" "extraTrees"
## [49] "fda" "FH.GBML" "FIR.DM"
## [52] "foba" "FRBCS.CHI" "FRBCS.W"
## [55] "FS.HGD" "gam" "gamboost"
## [58] "gamLoess" "gamSpline" "gaussprLinear"
## [61] "gaussprPoly" "gaussprRadial" "gbm_h2o"
## [64] "gbm" "gcvEarth" "GFS.FR.MOGUL"
## [67] "GFS.LT.RS" "GFS.THRIFT" "glm.nb"
## [70] "glm" "glmboost" "glmnet_h2o"
## [73] "glmnet" "glmStepAIC" "gpls"
## [76] "hda" "hdda" "hdrda"
## [79] "HYFIS" "icr" "J48"
## [82] "JRip" "kernelpls" "kknn"
## [85] "knn" "krlsPoly" "krlsRadial"
## [88] "lars" "lars2" "lasso"
## [91] "lda" "lda2" "leapBackward"
## [94] "leapForward" "leapSeq" "Linda"
## [97] "lm" "lmStepAIC" "LMT"
## [100] "loclda" "logicBag" "LogitBoost"
## [103] "logreg" "lssvmLinear" "lssvmPoly"
## [106] "lssvmRadial" "lvq" "M5"
## [109] "M5Rules" "manb" "mda"
## [112] "Mlda" "mlp" "mlpKerasDecay"
## [115] "mlpKerasDecayCost" "mlpKerasDropout" "mlpKerasDropoutCost"
## [118] "mlpML" "mlpSGD" "mlpWeightDecay"
## [121] "mlpWeightDecayML" "monmlp" "msaenet"
## [124] "multinom" "mxnet" "mxnetAdam"
## [127] "naive_bayes" "nb" "nbDiscrete"
## [130] "nbSearch" "neuralnet" "nnet"
## [133] "nnls" "nodeHarvest" "null"
## [136] "OneR" "ordinalNet" "ordinalRF"
## [139] "ORFlog" "ORFpls" "ORFridge"
## [142] "ORFsvm" "ownn" "pam"
## [145] "parRF" "PART" "partDSA"
## [148] "pcaNNet" "pcr" "pda"
## [151] "pda2" "penalized" "PenalizedLDA"
## [154] "plr" "pls" "plsRglm"
## [157] "polr" "ppr" "pre"
## [160] "PRIM" "protoclass" "qda"
## [163] "QdaCov" "qrf" "qrnn"
## [166] "randomGLM" "ranger" "rbf"
## [169] "rbfDDA" "Rborist" "rda"
## [172] "regLogistic" "relaxo" "rf"
## [175] "rFerns" "RFlda" "rfRules"
## [178] "ridge" "rlda" "rlm"
## [181] "rmda" "rocc" "rotationForest"
## [184] "rotationForestCp" "rpart" "rpart1SE"
## [187] "rpart2" "rpartCost" "rpartScore"
## [190] "rqlasso" "rqnc" "RRF"
## [193] "RRFglobal" "rrlda" "RSimca"
## [196] "rvmLinear" "rvmPoly" "rvmRadial"
## [199] "SBC" "sda" "sdwd"
## [202] "simpls" "SLAVE" "slda"
## [205] "smda" "snn" "sparseLDA"
## [208] "spikeslab" "spls" "stepLDA"
## [211] "stepQDA" "superpc" "svmBoundrangeString"
## [214] "svmExpoString" "svmLinear" "svmLinear2"
## [217] "svmLinear3" "svmLinearWeights" "svmLinearWeights2"
## [220] "svmPoly" "svmRadial" "svmRadialCost"
## [223] "svmRadialSigma" "svmRadialWeights" "svmSpectrumString"
## [226] "tan" "tanSearch" "treebag"
## [229] "vbmpRadial" "vglmAdjCat" "vglmContRatio"
## [232] "vglmCumulative" "widekernelpls" "WM"
## [235] "wsrf" "xgbDART" "xgbLinear"
## [238] "xgbTree" "xyf"set.seed(42)
x <- matrix(rnorm(50*5), ncol=5)
colnames(x) <- paste0("V", 1:5)
y <- factor(rep(c("A", "B"), 25))
featurePlot(x = x, y = y, plot = "box",
scales = list(y = list(relation="free"), x = list(rot = 90)),
layout = c(5,1))
featurePlot(x = x, y = y, plot = "density",
scales = list(x = list(relation="free"), y = list(relation="free")),
auto.key = list(columns = 2))
featurePlot(x = x, y = y, plot = "pairs",
auto.key = list(columns = 2))
featurePlot(x = x, y = y, plot = "ellipse",
auto.key = list(columns = 2))
Вывод: x и y создавались отдельно друг от друга (rnorm() никак не привязан к классам), поэтому на графиках классы A и B перекрываются везде — по этим признакам их не отличить. Так и должно быть, раз данные случайные.
Задание 2. Важность признаков через FSelector (data(iris))
Оценим важность признаков для классификации ирисов тремя разными метриками: information gain, gain ratio и chi-squared.
library(FSelector)
data(iris)
weights_ig <- information.gain(Species~., iris)
print(weights_ig)## attr_importance
## Sepal.Length 0.4521286
## Sepal.Width 0.2672750
## Petal.Length 0.9402853
## Petal.Width 0.9554360subset_ig <- cutoff.k(weights_ig, 2)
print(subset_ig)## [1] "Petal.Width" "Petal.Length"weights_gr <- gain.ratio(Species~., iris)
print(weights_gr)## attr_importance
## Sepal.Length 0.4196464
## Sepal.Width 0.2472972
## Petal.Length 0.8584937
## Petal.Width 0.8713692weights_chi <- chi.squared(Species~., iris)
print(weights_chi)## attr_importance
## Sepal.Length 0.6288067
## Sepal.Width 0.4922162
## Petal.Length 0.9346311
## Petal.Width 0.9432359comparison <- data.frame(
Feature = rownames(weights_ig),
InfoGain = weights_ig$attr_importance,
GainRatio = weights_gr$attr_importance,
ChiSquared = weights_chi$attr_importance
)
print(comparison)## Feature InfoGain GainRatio ChiSquared
## 1 Sepal.Length 0.4521286 0.4196464 0.6288067
## 2 Sepal.Width 0.2672750 0.2472972 0.4922162
## 3 Petal.Length 0.9402853 0.8584937 0.9346311
## 4 Petal.Width 0.9554360 0.8713692 0.9432359Задание 3. Дискретизация переменных: discretize() из пакета arules
Преобразуем непрерывный признак Sepal.Length набора iris в категориальный четырьмя разными способами.
library(arules)## Загрузка требуемого пакета: Matrix##
## Присоединяю пакет: 'arules'## Следующие объекты скрыты от 'package:base':
##
## abbreviate, writedata(iris)
x <- iris$Sepal.Length
hist(x, breaks = 20, main = "Исходное распределение Sepal.Length")
# Метод "interval" — интервалы равной ширины
disc_interval <- discretize(x, method = "interval", breaks = 3)
table(disc_interval)## disc_interval
## [4.3,5.5) [5.5,6.7) [6.7,7.9]
## 52 70 28# Метод "frequency" — интервалы равной частоты (равное число наблюдений)
disc_frequency <- discretize(x, method = "frequency", breaks = 3)
table(disc_frequency)## disc_frequency
## [4.3,5.4) [5.4,6.3) [6.3,7.9]
## 46 53 51# Метод "cluster" — границы через кластеризацию k-means
disc_cluster <- discretize(x, method = "cluster", breaks = 3)
table(disc_cluster)## disc_cluster
## [4.3,5.33) [5.33,6.27) [6.27,7.9]
## 46 53 51# Метод "fixed" — границы задаются вручную
disc_fixed <- discretize(x, method = "fixed",
breaks = c(-Inf, 5.5, 6.5, Inf),
labels = c("short", "medium", "long"))
table(disc_fixed)## disc_fixed
## short medium long
## 52 63 35# Сравнение всех четырёх методов в одной таблице
comparison_disc <- data.frame(
Sepal.Length = x,
Interval = disc_interval,
Frequency = disc_frequency,
Cluster = disc_cluster,
Fixed = disc_fixed
)
head(comparison_disc, 10)## Sepal.Length Interval Frequency Cluster Fixed
## 1 5.1 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 2 4.9 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 3 4.7 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 4 4.6 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 5 5.0 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 6 5.4 [4.3,5.5) [5.4,6.3) [5.33,6.27) short
## 7 4.6 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 8 5.0 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 9 4.4 [4.3,5.5) [4.3,5.4) [4.3,5.33) short
## 10 4.9 [4.3,5.5) [4.3,5.4) [4.3,5.33) shortЗадание 4*. Отбор признаков алгоритмом Boruta (data(“Ozone”))
используем алгоритм Boruta, чтобы определить, какие признаки значимо влияют на концентрацию озона (V4), а какие — нет
library(Boruta)
library(mlbench)
data("Ozone")
ozone <- na.omit(Ozone)
summary(ozone)## V1 V2 V3 V4 V5
## 3 :21 9 : 9 1:37 Min. : 1.00 Min. :5320
## 4 :21 12 : 9 2:45 1st Qu.: 5.00 1st Qu.:5690
## 12 :21 13 : 8 3:43 Median : 9.00 Median :5760
## 10 :18 14 : 8 4:36 Mean :11.37 Mean :5746
## 1 :17 15 : 8 5:42 3rd Qu.:16.00 3rd Qu.:5830
## 2 :17 22 : 8 6: 0 Max. :38.00 Max. :5950
## (Other):88 (Other):153 7: 0
## V6 V7 V8 V9
## Min. : 0.000 Min. :19.00 Min. :25.00 Min. :27.68
## 1st Qu.: 3.000 1st Qu.:46.00 1st Qu.:51.50 1st Qu.:49.64
## Median : 5.000 Median :64.00 Median :61.00 Median :56.48
## Mean : 4.867 Mean :57.61 Mean :61.11 Mean :56.54
## 3rd Qu.: 6.000 3rd Qu.:73.00 3rd Qu.:71.00 3rd Qu.:66.20
## Max. :11.000 Max. :93.00 Max. :93.00 Max. :82.58
##
## V10 V11 V12 V13
## Min. : 111 Min. :-69.00 Min. :27.50 Min. : 0.0
## 1st Qu.: 869 1st Qu.:-14.00 1st Qu.:51.26 1st Qu.: 60.0
## Median :2083 Median : 18.00 Median :60.98 Median :100.0
## Mean :2602 Mean : 14.43 Mean :60.69 Mean :122.2
## 3rd Qu.:5000 3rd Qu.: 43.00 3rd Qu.:70.88 3rd Qu.:150.0
## Max. :5000 Max. :107.00 Max. :90.68 Max. :350.0
## set.seed(123)
Boruta.Ozone <- Boruta(V4 ~ ., data = ozone, doTrace = 0)
print(Boruta.Ozone)## Boruta performed 18 iterations in 0.261224 secs.
## 9 attributes confirmed important: V1, V10, V11, V12, V13 and 4 more;
## 3 attributes confirmed unimportant: V2, V3, V6;plot(Boruta.Ozone, las = 2, cex.axis = 0.7,
main = "Важность признаков по алгоритму Boruta (Ozone)")
Вывод: алгоритм Boruta делит все признаки на три группы: Confirmed (значимо важные — зелёные на графике), Rejected (незначимые — красные) и Tentative (не удалось однозначно определить — жёлтые). Boruta работает надёжнее, чем разовая оценка важности (например, через один Random Forest), потому что сравнивает каждый признак с его “теневой” перемешанной копией много раз подряд, отсекая случайные совпадения. Confirmed-признаки — это те, что стоит оставить в модели, Rejected — можно смело убрать без потери качества предсказания.