Random Forest

Krissie

2024-11-27

R Markdown

#1. Loading data

setwd("/Users/kitsanasudsaneh/Desktop/fall2024/MAX503/final Project")
wine.raw <- read.csv("vivino.csv")

#2.Inspect the dataset

#---Dimensions of the dataset

n_rows <- nrow (wine.raw) #Number of observations (rows)
n_columns <- ncol (wine.raw) #Number of variables (columns)

cat("The dataset contains: \n")

## The dataset contains:

cat(n_rows,"Oberservation (rows)\n")

## 1599 Oberservation (rows)

cat(n_columns,"Variables (column)\n")

## 12 Variables (column)

#---Sumarize the dataset

summary.wine <- summary(wine.raw)
cat("Summary of the dataset: \n")

## Summary of the dataset:

summary.wine

## fixed.acidity volatile.acidity citric.acid residual.sugar

## Min. : 4.60 Min. :0.1200 Min. :0.000 Min. : 0.900
## 1st Qu.: 7.10 1st Qu.:0.3900 1st Qu.:0.090 1st Qu.: 1.900
## Median : 7.90 Median :0.5200 Median :0.260 Median : 2.200
## Mean : 8.32 Mean :0.5278 Mean :0.271 Mean : 2.539

1
## 3rd Qu.: 9.20 3rd Qu.:0.6400 3rd Qu.:0.420 3rd Qu.: 2.600
## Max. :15.90 Max. :1.5800 Max. :1.000 Max. :15.500
## chlorides free.sulfur.dioxide total.sulfur.dioxide density
## Min. :0.01200 Min. : 1.00 Min. : 6.00 Min. :0.9901
## 1st Qu.:0.07000 1st Qu.: 7.00 1st Qu.: 22.00 1st Qu.:0.9956
## Median :0.07900 Median :14.00 Median : 38.00 Median :0.9968
## Mean :0.08747 Mean :15.87 Mean : 46.47 Mean :0.9967
## 3rd Qu.:0.09000 3rd Qu.:21.00 3rd Qu.: 62.00 3rd Qu.:0.9978
## Max. :0.61100 Max. :72.00 Max. :289.00 Max. :1.0037
## pH sulphates alcohol quality
## Min. :2.740 Min. :0.3300 Min. : 8.40 Min. :3.000
## 1st Qu.:3.210 1st Qu.:0.5500 1st Qu.: 9.50 1st Qu.:5.000
## Median :3.310 Median :0.6200 Median :10.20 Median :6.000
## Mean :3.311 Mean :0.6581 Mean :10.42 Mean :5.636
## 3rd Qu.:3.400 3rd Qu.:0.7300 3rd Qu.:11.10 3rd Qu.:6.000
## Max. :4.010 Max. :2.0000 Max. :14.90 Max. :8.000

#---Structure of the dataset

cat("Structure of the dataset: \n")

## Structure of the dataset:

str(wine.raw)

## ’data.frame’: 1599 obs. of 12 variables:

## $ fixed.acidity : num 7.4 7.8 7.8 11.2 7.4 7.4 7.9 7.3 7.8 7.5 ...
## $ volatile.acidity : num 0.7 0.88 0.76 0.28 0.7 0.66 0.6 0.65 0.58 0.5 ...
## $ citric.acid : num 0 0 0.04 0.56 0 0 0.06 0 0.02 0.36 ...
## $ residual.sugar : num 1.9 2.6 2.3 1.9 1.9 1.8 1.6 1.2 2 6.1 ...
## $ chlorides : num 0.076 0.098 0.092 0.075 0.076 0.075 0.069 0.065 0.073 0.071 ...
## $ free.sulfur.dioxide : num 11 25 15 17 11 13 15 15 9 17 ...
## $ total.sulfur.dioxide: num 34 67 54 60 34 40 59 21 18 102 ...
## $ density : num 0.998 0.997 0.997 0.998 0.998 ...
## $ pH : num 3.51 3.2 3.26 3.16 3.51 3.51 3.3 3.39 3.36 3.35 ...
## $ sulphates : num 0.56 0.68 0.65 0.58 0.56 0.56 0.46 0.47 0.57 0.8 ...
## $ alcohol : num 9.4 9.8 9.8 9.8 9.4 9.4 9.4 10 9.5 10.5 ...
## $ quality : int 5 5 5 6 5 5 5 7 7 5 ...

#---Correlation Plot
wine.df <- wine.raw[,-12]
#Exclude 'quality': group the data under the quality column

#compute correlation matrix:

library(corrplot)

## Warning: package ’corrplot’ was built under R version 4.3.3

## corrplot 0.95 loaded

corr_matrix <-cor(wine.df, use = "complete.obs")

corr_matrix

2
## fixed.acidity volatile.acidity citric.acid residual.sugar
## fixed.acidity 1.00000000 -0.256130895 0.67170343 0.114776724
## volatile.acidity -0.25613089 1.000000000 -0.55249568 0.001917882
## citric.acid 0.67170343 -0.552495685 1.00000000 0.143577162
## residual.sugar 0.11477672 0.001917882 0.14357716 1.000000000
## chlorides 0.09370519 0.061297772 0.20382291 0.055609535
## free.sulfur.dioxide -0.15379419 -0.010503827 -0.06097813 0.187048995
## total.sulfur.dioxide -0.11318144 0.076470005 0.03553302 0.203027882
## density 0.66804729 0.022026232 0.36494718 0.355283371
## pH -0.68297819 0.234937294 -0.54190414 -0.085652422
## sulphates 0.18300566 -0.260986685 0.31277004 0.005527121
## alcohol -0.06166827 -0.202288027 0.10990325 0.042075437
## chlorides free.sulfur.dioxide total.sulfur.dioxide
## fixed.acidity 0.093705186 -0.153794193 -0.11318144
## volatile.acidity 0.061297772 -0.010503827 0.07647000
## citric.acid 0.203822914 -0.060978129 0.03553302
## residual.sugar 0.055609535 0.187048995 0.20302788
## chlorides 1.000000000 0.005562147 0.04740047
## free.sulfur.dioxide 0.005562147 1.000000000 0.66766645
## total.sulfur.dioxide 0.047400468 0.667666450 1.00000000
## density 0.200632327 -0.021945831 0.07126948
## pH -0.265026131 0.070377499 -0.06649456
## sulphates 0.371260481 0.051657572 0.04294684
## alcohol -0.221140545 -0.069408354 -0.20565394
## density pH sulphates alcohol
## fixed.acidity 0.66804729 -0.68297819 0.183005664 -0.06166827
## volatile.acidity 0.02202623 0.23493729 -0.260986685 -0.20228803
## citric.acid 0.36494718 -0.54190414 0.312770044 0.10990325
## residual.sugar 0.35528337 -0.08565242 0.005527121 0.04207544
## chlorides 0.20063233 -0.26502613 0.371260481 -0.22114054
## free.sulfur.dioxide -0.02194583 0.07037750 0.051657572 -0.06940835
## total.sulfur.dioxide 0.07126948 -0.06649456 0.042946836 -0.20565394
## density 1.00000000 -0.34169933 0.148506412 -0.49617977
## pH -0.34169933 1.00000000 -0.196647602 0.20563251
## sulphates 0.14850641 -0.19664760 1.000000000 0.09359475
## alcohol -0.49617977 0.20563251 0.093594750 1.00000000

#plot the correltion:

corrplot(corr_matrix, method = "circle", type = "upper",
tl.cex = 0.8, tl.col = "black", addCoef.col = "black",
main = "\n\n\nPhysicochemical Properties of Bordeaux wine Correlation Plot")

3
 total.sulfur.dioxide
free.sulfur.dioxide
residual.sugar
volatile.acidity
Physicochemical Properties of Bordeaux wine Correlation Plot

fixed.acidity

sulphates
citric.acid

chlorides

density

alcohol
pH
1
fixed.acidity 1.00
−0.260.670.110.09
−0.15
−0.110.67
−0.680.18
−0.06
0.8
volatile.acidity 1.00
−0.550.000.06
−0.010.080.020.23
−0.26
−0.20
0.6
citric.acid 1.000.140.20
−0.060.040.36
−0.540.310.11
residual.sugar 1.000.060.190.200.36
−0.090.010.04 0.4

chlorides 1.000.010.050.20 −0.22 0.2

−0.270.37
free.sulfur.dioxide 1.000.67
−0.020.070.05
−0.07 0

total.sulfur.dioxide 1.000.07
−0.070.04
−0.21 −0.2
density 1.00
−0.340.15
−0.50 −0.4
pH 1.00−0.200.21 −0.6
sulphates 1.000.09
−0.8
alcohol 1.00
−1

#3.Cleaning and Preparing the data

#---Create a Naw Column with two groups(bad and good)

wine.raw$quality.label <-ifelse(wine.raw$quality > 6, "Good","Bad")
print(head(wine.raw$quality.label))

## [1] "Bad" "Bad" "Bad" "Bad" "Bad" "Bad"

table(wine.raw$quality.label)

##
## Bad Good
## 1382 217

head(wine.raw)

## fixed.acidity volatile.acidity citric.acid residual.sugar chlorides

## 1 7.4 0.70 0.00 1.9 0.076
## 2 7.8 0.88 0.00 2.6 0.098
## 3 7.8 0.76 0.04 2.3 0.092
## 4 11.2 0.28 0.56 1.9 0.075
## 5 7.4 0.70 0.00 1.9 0.076
## 6 7.4 0.66 0.00 1.8 0.075
## free.sulfur.dioxide total.sulfur.dioxide density pH sulphates alcohol

4
## 1 11 34 0.9978 3.51 0.56 9.4
## 2 25 67 0.9968 3.20 0.68 9.8
## 3 15 54 0.9970 3.26 0.65 9.8
## 4 17 60 0.9980 3.16 0.58 9.8
## 5 11 34 0.9978 3.51 0.56 9.4
## 6 13 40 0.9978 3.51 0.56 9.4
## quality quality.label
## 1 5 Bad
## 2 5 Bad
## 3 5 Bad
## 4 6 Bad
## 5 5 Bad
## 6 5 Bad

#---Splitting Data into Training and Testing Sets

set.seed(04625)
train.prop <- 0.70 #70% training, 30% testing
train.cases <- sample(nrow(wine.raw), nrow(wine.raw) * train.prop)
train.cases

## [1] 360 429 591 831 1108 350 1138 1127 1477 31 869 529 842 327
## [15] 239 678 1497 1057 230 87 608 399 977 232 286 1136 471 486
## [29] 705 1455 473 154 1088 57 1140 619 806 1052 582 1114 930 1401
## [43] 1560 878 982 1448 249 1244 504 521 1080 148 1333 1503 382 994
## [57] 534 1226 1087 646 255 1469 70 309 159 1273 737 256 1089 143
## [71] 246 225 207 1517 361 1433 1579 43 613 1451 400 433 901 1397
## [85] 459 729 155 233 85 115 1308 579 338 880 170 1188 1212 689
## [99] 622 1202 1269 45 570 750 587 557 601 250 1093 1494 1305 1432
## [113] 947 1518 541 27 1042 709 223 60 710 1192 369 1468 291 1535
## [127] 1291 1 510 1032 1169 1153 183 563 1224 1383 941 1250 1354 417
## [141] 348 943 359 1373 1381 54 766 1464 224 1148 38 1293 659 317
## [155] 1595 687 1123 1067 824 1098 1151 735 387 90 547 1006 1105 986
## [169] 100 518 1191 390 1589 1003 826 1558 685 667 192 108 193 1454
## [183] 961 1453 1068 1133 1499 912 762 1144 823 1425 1315 1320 817 320
## [197] 712 370 453 310 1211 1566 857 336 380 1241 1440 141 1311 11
## [211] 1427 1498 716 1084 1475 35 1073 1272 1116 1015 421 684 408 307
## [225] 1520 791 852 74 26 820 850 1312 173 515 500 1394 630 662
## [239] 535 959 927 796 337 383 103 1081 1339 1100 32 460 810 1221
## [253] 565 606 341 968 28 413 1360 558 507 1165 631 1036 1322 180
## [267] 278 23 29 536 424 1059 954 248 396 264 161 1287 1082 1387
## [281] 254 756 1172 358 1078 458 844 477 780 372 877 995 538 924
## [295] 328 435 1086 438 1150 1450 1280 1235 283 938 765 945 1259 321
## [309] 71 99 568 15 907 479 663 1024 543 1157 1160 1069 922 411
## [323] 1090 1205 502 1332 67 990 1438 829 1301 102 185 1121 128 105
## [337] 1565 1023 751 733 1183 69 1353 949 24 861 1298 726 292 539
## [351] 293 404 53 395 940 1079 195 420 40 1213 322 957 1483 2
## [365] 1178 879 1285 862 1361 993 371 1118 655 1393 594 306 1594 1257
## [379] 80 795 1139 1545 569 1508 497 1430 524 713 926 599 1255 178
## [393] 523 1302 586 1135 566 326 1045 279 1346 653 1487 316 1245 186
## [407] 86 813 1562 456 1268 1474 632 189 648 1179 323 257 1486 245
## [421] 1016 618 1395 138 969 723 1247 1573 1058 491 485 635 120 537
## [435] 349 1142 346 998 1553 1209 285 1203 643 715 1352 260 335 1584
## [449] 866 49 126 1227 1435 97 449 1480 34 894 1400 68 3 1130
## [463] 325 1296 992 190 512 1230 1403 140 163 149 770 690 445 1107

5
## [477] 1509 373 1532 564 1412 258 1356 1066 987 785 1033 794 778 6
## [491] 1225 20 873 683 481 1345 624 123 58 574 1194 302 137 527
## [505] 1330 719 1514 1367 1472 480 1347 450 609 1369 581 944 1364 98
## [519] 454 1375 983 1064 533 1115 848 288 275 895 634 833 1005 1587
## [533] 540 984 84 644 525 989 1094 871 244 808 585 747 1208 1569
## [547] 1031 962 846 1270 812 867 1261 1220 1349 150 1419 131 775 61
## [561] 802 617 657 860 929 1362 1274 682 393 4 1129 718 410 980
## [575] 967 1263 104 1557 1335 658 1001 1174 991 1318 893 273 281 626
## [589] 1544 654 876 332 800 1132 1465 432 59 830 1456 437 1223 607
## [603] 707 1546 18 628 211 1404 935 1590 671 739 681 1189 478 375
## [617] 295 647 164 1041 1398 517 670 1051 722 845 1228 1054 641 872
## [631] 891 1488 1161 516 550 748 113 129 472 1409 827 639 532 144
## [645] 567 1442 253 779 1334 1104 700 1452 905 76 56 401 948 381
## [659] 378 908 978 767 55 1176 151 666 546 714 466 577 496 1193
## [673] 467 237 1017 633 1458 196 819 837 1038 571 252 1251 259 236
## [687] 915 973 240 475 1444 1149 1522 1555 1239 1384 637 677 1526 1567
## [701] 1175 1543 355 1530 1292 1538 1237 1185 1365 1267 296 112 1170 1554
## [715] 1145 175 1598 1206 19 971 93 202 669 706 476 673 815 702
## [729] 1571 580 427 9 1559 1583 1278 711 548 482 717 625 906 1408
## [743] 886 367 452 443 36 134 612 593 168 651 1187 828 1277 483
## [757] 953 50 1158 1025 299 832 352 782 1040 909 1309 106 772 976
## [771] 14 996 1493 821 228 605 201 809 441 855 1495 1216 1290 1420
## [785] 1249 167 499 1266 562 470 522 1037 1577 1252 1204 305 1529 1026
## [799] 610 1011 206 402 124 1162 874 1556 166 16 1462 199 970 732
## [813] 741 1576 1460 1531 981 1184 1218 636 725 597 1549 892 1338 263
## [827] 1541 1447 13 788 447 203 703 474 52 1101 268 660 1463 847
## [841] 234 1313 446 132 545 514 1177 692 1217 900 362 51 364 590
## [855] 210 738 1146 549 925 1441 531 78 1491 629 1207 389 484 966
## [869] 1106 985 488 1119 542 1126 884 1195 856 864 1421 1446 1378 870
## [883] 578 315 934 675 160 1271 955 181 490 665 10 1168 1186 530
## [897] 781 91 1593 572 958 312 1070 1128 300 790 1248 724 721 48
## [911] 1143 965 875 801 135 863 147 271 1022 95 919 461 544 1382
## [925] 218 374 79 1306 917 448 600 825 1467 1336 388 1152 47 1007
## [939] 415 814 1197 1550 776 503 1540 270 229 933 792 950 627 640
## [953] 1578 695 386 1295 351 1426 133 761 734 130 66 1570 889 885
## [967] 588 1310 266 414 645 1200 854 290 153 743 674 385 1370 469
## [981] 804 1234 951 122 1008 1340 556 730 708 1428 1337 805 1390 759
## [995] 1406 220 304 1363 1479 736 693 1085 114 379 462 1159 1163 267
## [1009] 1283 1300 22 1344 511 1021 1449 1095 1473 1122 284 1399 1439 1489
## [1023] 274 227 1281 287 25 952 928 745 405 960 1396 493 1415 997
## [1037] 354 1359 1096 261 1484 1459 946 799 1470 194 1062 1043 276 554
## [1051] 789 528 686 834 903 92 701 1581 1591 598 904 787 208 226
## [1065] 1262 1260 1358 1155 1582 188 918 868 1265 742 1515 931 94 1355
## [1079] 1551 818 1060 1125 881 1501 691 584 1233 265 440 1478 798 1342
## [1093] 744 1506 1156 757 182 899 1552 282 157 1083 505 162 592 1507
## [1107] 1376 174 125 1034 975 1020 939 212 890 1536 465 1099 1167

wine.df.train <- wine.raw[train.cases,]

wine.df.test <- wine.raw[-train.cases,]

#---Checking the detail of sampling

cat("Training data dimensions:", dim(wine.df.train), "\n")

## Training data dimensions: 1119 13

6
 print(table(wine.df.train$quality.label))

##
## Bad Good
## 956 163

#>> The training data set contains 956 wines labeled as "Bad" and
#>> 163 wines labeled as "Good": model will learn patterns from these sampling
cat("Testing data dimensions :", dim(wine.df.test), "\n")

## Testing data dimensions : 480 13

print(table(wine.df.test$quality.label))

##
## Bad Good
## 426 54

#>> The testing data set contains 426 wines labeled as "Bad" and
#>> 54 wines labeled as "Good": not seen these sampling

###--------------------------------------------------------------------- ###
##-----------------------------Random Forest--------------------------##

#---Ensure the target variable is a factor

wine.df.train$quality.label <-as.factor(wine.df.train$quality.label)
wine.df.test$quality.label <- as.factor(wine.df.test$quality.label)

#4.Building a Random Forest Model:

library(randomForest)

## Warning: package ’randomForest’ was built under R version 4.3.3

## randomForest 4.7-1.2

## Type rfNews() to see new features/changes/bug fixes.

set.seed(123)
wine.rf <- randomForest(quality.label ~.-quality, #all features except quality
data = wine.df.train, ntree = 3000)
wine.rf

##
## Call:
## randomForest(formula = quality.label ~ . - quality, data = wine.df.train, ntree = 3000)
## Type of random forest: classification
## Number of trees: 3000
## No. of variables tried at each split: 3
##

7
## OOB estimate of error rate: 8.94%
## Confusion matrix:
## Bad Good class.error
## Bad 931 25 0.02615063
## Good 75 88 0.46012270

#5.Making Predictions for the test data:

wine.rf.predict <- predict(wine.rf,wine.df.test)
wine.rf.predict

## 5 7 8 12 17 21 30 33 37 39 41 42 44 46 62 63
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 64 65 72 73 75 77 81 82 83 88 89 96 101 107 109 110
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 111 116 117 118 119 121 127 136 139 142 145 146 152 156 158 165
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 169 171 172 176 177 179 184 187 191 197 198 200 204 205 209 213
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 214 215 216 217 219 221 222 231 235 238 241 242 243 247 251 262
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 269 272 277 280 289 294 297 298 301 303 308 311 313 314 318 319
## Bad Bad Bad Good Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good
## 324 329 330 331 333 334 339 340 342 343 344 345 347 353 356 357
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 363 365 366 368 376 377 384 391 392 394 397 398 403 406 407 409
## Bad Good Good Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Good
## 412 416 418 419 422 423 425 426 428 430 431 434 436 439 442 444
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Bad
## 451 455 457 463 464 468 487 489 492 494 495 498 501 506 508 509
## Bad Bad Bad Bad Bad Good Bad Bad Good Bad Bad Bad Bad Good Bad Bad
## 513 519 520 526 551 552 553 555 559 560 561 573 575 576 583 589
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 595 596 602 603 604 611 614 615 616 620 621 623 638 642 649 650
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 652 656 661 664 668 672 676 679 680 688 694 696 697 698 699 704
## Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 720 727 728 731 740 746 749 752 753 754 755 758 760 763 764 768
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 769 771 773 774 777 783 784 786 793 797 803 807 811 816 822 835
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Good Bad Bad Bad Bad
## 836 838 839 840 841 843 849 851 853 858 859 865 882 883 887 888
## Bad Good Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad
## 896 897 898 902 910 911 913 914 916 920 921 923 932 936 937 942
## Bad Good Bad Good Bad Bad Bad Good Good Bad Bad Bad Bad Bad Bad Bad
## 956 963 964 972 974 979 988 999 1000 1002 1004 1009 1010 1012 1013 1014
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Good Bad Bad Bad Bad
## 1018 1019 1027 1028 1029 1030 1035 1039 1044 1046 1047 1048 1049 1050 1053 1055
## Bad Bad Bad Bad Bad Good Bad Good Bad Bad Bad Bad Bad Bad Bad Bad
## 1056 1061 1063 1065 1071 1072 1074 1075 1076 1077 1091 1092 1097 1102 1103 1109
## Bad Bad Bad Bad Good Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad
## 1110 1111 1112 1113 1117 1120 1124 1131 1134 1137 1141 1147 1154 1164 1166 1171
## Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good
## 1173 1180 1181 1182 1190 1196 1198 1199 1201 1210 1214 1215 1219 1222 1229 1231
## Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good

8
## 1232 1236 1238 1240 1242 1243 1246 1253 1254 1256 1258 1264 1275 1276 1279 1282
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1284 1286 1288 1289 1294 1297 1299 1303 1304 1307 1314 1316 1317 1319 1321 1323
## Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Good
## 1324 1325 1326 1327 1328 1329 1331 1341 1343 1348 1350 1351 1357 1366 1368 1371
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1372 1374 1377 1379 1380 1385 1386 1388 1389 1391 1392 1402 1405 1407 1410 1411
## Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad
## 1413 1414 1416 1417 1418 1422 1423 1424 1429 1431 1434 1436 1437 1443 1445 1457
## Good Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1461 1466 1471 1476 1481 1482 1485 1490 1492 1496 1500 1502 1504 1505 1510 1511
## Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad
## 1512 1513 1516 1519 1521 1523 1524 1525 1527 1528 1533 1534 1537 1539 1542 1547
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1548 1561 1563 1564 1568 1572 1574 1575 1580 1585 1586 1588 1592 1596 1597 1599
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## Levels: Bad Good

#6.Visualizing Predictions:
library(cluster)
clusplot(wine.df.test[-13], wine.rf.predict, color = TRUE, shade = TRUE,
labels = 2, lines = 0,
main ="Cluster Plot of Random Forest Predictions")

Cluster Plot of Random Forest Predictions

589 Good Bad

1004 807
4

492
468
822 897 942
914 506431
1476 1210
910
1418 916
1009
11731077
839
911
1071
444 841
1091
972 376 340
1229
145 1027
979 1585
858
1113 913
1303 859
519
342 1061
391
1000
803 1243
1231
1092 1324
1063
1039
1505
1510
649
1542 1372
1214 664
331
366409442
2

696 1299
13911112
1288
838 11021019
1018
1504
1434 1323
937
936 883
1171
1049
1182
406 1137
1215
888
1407
1413
964
1002
1050 463
213
1222 198
272
242
1238
2311574
1120 1490
3471065
1572
103011341586
974
1599 1181
1180
1219
1012 1124
956
1154
1199 407
384 344
343
251 419
560 329
451
1317 96 1548
1053
1492 1103
15008 1496
1528
1328
1327
1326
1325
902 294
1198
1201
1111 289 280
1482
31917 1044
573 489
559668
377
345
1417
356 37
1275 63
1131 1076
1147 774
116 398
561 576 392
Component 2

1592 1046
1588 455661
172614 513 365
461411 426
422
1539
1580
1341
1357
1380
1117
1597
169
1513
1537
882
1596 551
1521
1379
920
923
1533
1525
811
1343
30749
746
1511
1314
1282
1527
301
1423763
81
1431
118
269
277
1519
1405
1304
205
117
1074
119
1523
999
101
1258
1392 44
221
428 2177
921
552
1368
851
457
553
109
676 843
816 311
1286
330
680
308
620
508
439
403
575
1110 436
434
430
363 339
602
604
755
3571166 555
1410 495
698
697
1246 88
177
887 487
0

896
8981196
1097
728
1416
1547
65 219
1461 334
1029
1568
1564
1563
1254963
1014
784
1048 501
217
727
215
2041284
111
1471
494
1164 623 786583
1010 615
368
1457
1240
1256
1481 840
1350
1348
849
1253
10471388
1524
197
238
1329 1
1377
1445
298176 672
777
1141
331
1316
1319
353 1056
1055
187
764
1035
1366
1389520 41
313 509
1414
611
1242 75 241731 82
107
10281485 1429
1443
932
1424
1512
179
425
423 793 1422
214835
1279
1276
303
1109
235 142
650
498
222
83 1575
1321
688
12
209 797
324 89
127 184
769
771
247
1294 7
5758
1466
603
64
1013988
139
1516
595 136
243318
1351
752
754333
1534 412
642
704 853
1289
216
616
1307 394
656 1371
200 171 836
720
865
1190 73753
72740158
156
1236
33621
768 699
62526
297 1437
694
191
42
760 152
−2

418
39 1232 1385
1264 1502
1561 314
773 464
262 783
1402
1374 110
1386 146
679
121
1297 165 652 596
638
1072
1075 416
−4

1436
397
−6

−5 0 5

Component 1
These two components explain 44.42 % of the point variability.

9
#7.Handing Imbalanced Classes

#---Building a Random Forest Model: with Balanced Sampling

#the model is now forced to focus equally on both classes.

library(randomForest)
set.seed(123)
wine.rf <- randomForest(quality.label ~.-quality, #all features except quality
data = wine.df.train, ntree = 3000,
sampsize = c(163,163)) #Match the smaller class size
wine.rf

##
## Call:
## randomForest(formula = quality.label ~ . - quality, data = wine.df.train, ntree = 3000, sampsiz
## Type of random forest: classification
## Number of trees: 3000
## No. of variables tried at each split: 3
##
## OOB estimate of error rate: 14.21%
## Confusion matrix:
## Bad Good class.error
## Bad 830 126 0.1317992
## Good 33 130 0.2024540

#---Making Predictions for the test data:

wine.rf.predict <- predict(wine.rf,wine.df.test)
wine.rf.predict

## 5 7 8 12 17 21 30 33 37 39 41 42 44 46 62 63
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 64 65 72 73 75 77 81 82 83 88 89 96 101 107 109 110
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 111 116 117 118 119 121 127 136 139 142 145 146 152 156 158 165
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 169 171 172 176 177 179 184 187 191 197 198 200 204 205 209 213
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Good
## 214 215 216 217 219 221 222 231 235 238 241 242 243 247 251 262
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Good Bad
## 269 272 277 280 289 294 297 298 301 303 308 311 313 314 318 319
## Bad Bad Bad Good Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good
## 324 329 330 331 333 334 339 340 342 343 344 345 347 353 356 357
## Bad Bad Bad Good Bad Bad Good Good Good Bad Bad Bad Good Bad Bad Good
## 363 365 366 368 376 377 384 391 392 394 397 398 403 406 407 409
## Bad Good Good Bad Good Bad Bad Good Good Bad Bad Good Bad Good Good Good
## 412 416 418 419 422 423 425 426 428 430 431 434 436 439 442 444
## Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Good Good
## 451 455 457 463 464 468 487 489 492 494 495 498 501 506 508 509
## Bad Bad Bad Good Bad Good Bad Good Good Bad Bad Bad Bad Good Bad Bad
## 513 519 520 526 551 552 553 555 559 560 561 573 575 576 583 589
## Bad Good Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Good Bad Good
## 595 596 602 603 604 611 614 615 616 620 621 623 638 642 649 650
## Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad

10
## 652 656 661 664 668 672 676 679 680 688 694 696 697 698 699 704
## Bad Bad Bad Good Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 720 727 728 731 740 746 749 752 753 754 755 758 760 763 764 768
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 769 771 773 774 777 783 784 786 793 797 803 807 811 816 822 835
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Good Bad Bad Good Bad
## 836 838 839 840 841 843 849 851 853 858 859 865 882 883 887 888
## Bad Good Good Bad Good Bad Bad Bad Bad Good Good Bad Bad Good Bad Good
## 896 897 898 902 910 911 913 914 916 920 921 923 932 936 937 942
## Bad Good Bad Good Good Good Good Good Good Good Bad Good Bad Good Good Good
## 956 963 964 972 974 979 988 999 1000 1002 1004 1009 1010 1012 1013 1014
## Good Bad Good Good Bad Bad Bad Bad Bad Good Good Good Bad Bad Bad Bad
## 1018 1019 1027 1028 1029 1030 1035 1039 1044 1046 1047 1048 1049 1050 1053 1055
## Good Good Good Bad Bad Good Bad Good Good Bad Bad Bad Good Good Bad Bad
## 1056 1061 1063 1065 1071 1072 1074 1075 1076 1077 1091 1092 1097 1102 1103 1109
## Bad Good Good Bad Good Bad Bad Bad Bad Good Good Good Bad Bad Bad Bad
## 1110 1111 1112 1113 1117 1120 1124 1131 1134 1137 1141 1147 1154 1164 1166 1171
## Bad Bad Bad Good Bad Good Good Bad Good Good Bad Bad Good Bad Bad Good
## 1173 1180 1181 1182 1190 1196 1198 1199 1201 1210 1214 1215 1219 1222 1229 1231
## Good Good Good Good Bad Bad Bad Bad Bad Good Good Bad Good Good Good Good
## 1232 1236 1238 1240 1242 1243 1246 1253 1254 1256 1258 1264 1275 1276 1279 1282
## Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1284 1286 1288 1289 1294 1297 1299 1303 1304 1307 1314 1316 1317 1319 1321 1323
## Bad Bad Bad Bad Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Good
## 1324 1325 1326 1327 1328 1329 1331 1341 1343 1348 1350 1351 1357 1366 1368 1371
## Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1372 1374 1377 1379 1380 1385 1386 1388 1389 1391 1392 1402 1405 1407 1410 1411
## Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad Bad
## 1413 1414 1416 1417 1418 1422 1423 1424 1429 1431 1434 1436 1437 1443 1445 1457
## Good Bad Bad Bad Good Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad
## 1461 1466 1471 1476 1481 1482 1485 1490 1492 1496 1500 1502 1504 1505 1510 1511
## Bad Bad Bad Good Bad Good Bad Good Bad Bad Bad Bad Bad Bad Good Bad
## 1512 1513 1516 1519 1521 1523 1524 1525 1527 1528 1533 1534 1537 1539 1542 1547
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Bad
## 1548 1561 1563 1564 1568 1572 1574 1575 1580 1585 1586 1588 1592 1596 1597 1599
## Bad Bad Bad Bad Bad Bad Bad Bad Bad Good Good Bad Bad Good Bad Good
## Levels: Bad Good

#---Visualizing Predictions:
library(cluster)
clusplot(wine.df.test[-13], wine.rf.predict, color = TRUE, shade = TRUE,
labels = 2, lines = 0,
main ="Cluster Plot of Random Forest Predictions
with balance classes")

11
 Cluster Plot of Random Forest Predictions
with balance classes
Good
589 1004 807
4

492
468
822 897 942
914 506 431
1476 Bad
1210
910 916911
10091077
839
1071
1173
444 841
1091
972 376 340
1229
145 1027 1418
9791585
858
1113 913
1303 859
519
342
391 1243
1231
1092 1324
1063
1039
1505 1372 331
664
366 1061
6961000
803 1510
649
1542 1214 409442
2

12991112
1391 1288
838 11021019
1018
1504
1434 1323 883
1171
1049
1182
937
936 406 1137
1215
888
1407
1413
964
1002 463
213
1222 198
242272
1238 1490
2311574
3471065
1572 11341586
974
1599 1050
1181
1180
1219
1012 1124
956
1154 407 344419
343 560 329
1317 1120 1103
1500 1030
1496
1528 294
1198
1201 1199
289 384
280
1482 251
1044
573 559451
489
668
377
345
Component 2

96 1548
1053
1492
1592356 8 1328
1327
1326
1325
902
1046
1588 1111
127537
45563
1131
661
172 319
1076
1147
614 17
774
116 1417
398
561 576
513 392
365
461411426
422
1539
1580
1341
1357
1380
1117
1597
169
1513
1537
882 551
1521
1423
1596
1379
920
923
1533
1525
811
134330749
746
1511
1314
1282
1527
1392
301
763
81
1431
118
269
277
1519
1405
1304
205
117
1074
119
1523
999
101
1258 44
221
428
77
921
21552
1368
851553
109
676843
816680
308
311
1286
330 620
508
575
1110439
403
430436
434
363 339
602
604755
1410 495
698
697 88
177
887 457
487 357 1166 555
0

896
898 1246
1196
1097
728
1416
1547
1568
1564
1563
1254 219
1461
963 334
727
1029 501
217
215
2041284
111
1471
494
1164 623786583
1010 615
368
1457
1240
1256
1481 65
1014
840
1350784
1048
1348
849
1253
10471388
1524
238
1329
1445 1
1377
176 672
777
1141
197331
1316
1319 1056
1055
187
764
1035
1366
52041
313509
1414
611
1242 75 241731 82 107
10281485 298
1429
1443
932
1424
1512
179 793 353
1389
1422
835
1279
1276
303
214 142
650
498
222
83 1575
797
1321 324 89
127 425
423 1
1466
769
771 1109
235
84
7
5758
603
64
247
1294
1013988 136
243318
1351
333
752
754
1534 688
12
209
412
642
704 853
1289
216
616394
656 1371
200 836
720
865
1190 139
1516
595
753
72
73 740
331307
158
156
1236
621
768
42 62526
699
297 1437
694
191 152
760
−2

171
418
39 1232 1385
1264 1502
1561 314 464
783
1402 773146
262 679
121
1297 1374 110
1386
165 652 596
638
1072
1075 416
−4

1436
397
−6

−5 0 5 10

Component 1
These two components explain 44.42 % of the point variability.

#8.Evaluating the model:How well did we predict the test data?

#---Accuracy Method: which is the proportion pf correct predictions.

mean(wine.df.test$quality.label == wine.rf.predict)

## [1] 0.8291667

#---Adjust Rand Index (ARI)

library(mclust)

## Warning: package ’mclust’ was built under R version 4.3.2

## Package ’mclust’ version 6.1.1

## Type ’citation("mclust")’ for citing this R package in publications.

adjustedRandIndex(wine.df.test$quality.label, wine.rf.predict)

## [1] 0.3205036

#---Confusion Matrix
table(wine.rf.predict, wine.df.test$quality.label)

12
##
## wine.rf.predict Bad Good
## Bad 355 11
## Good 71 43

#---Summary Functions:
wine.summ <- function(data, groups)
{
aggregate(data, list(groups), function(x) mean(as.numeric(x)))
}
wine.summ(wine.df.test, wine.rf.predict) #proposed quality

## Group.1 fixed.acidity volatile.acidity citric.acid residual.sugar chlorides

## 1 Bad 8.009836 0.5739754 0.2312568 2.518852 0.09119126
## 2 Good 9.085965 0.3892105 0.3965789 2.544737 0.07586842
## free.sulfur.dioxide total.sulfur.dioxide density pH sulphates
## 1 16.92213 52.24454 0.9968152 3.328743 0.6346175
## 2 13.64035 35.09649 0.9961684 3.275263 0.7420175
## alcohol quality quality.label
## 1 10.15574 5.434426 1.030055
## 2 11.47515 6.315789 1.377193

wine.summ(wine.df.test, wine.df.test$quality.label) #actual quality

## Group.1 fixed.acidity volatile.acidity citric.acid residual.sugar chlorides

## 1 Bad 8.261737 0.5419953 0.2649765 2.498944 0.08976761
## 2 Good 8.294444 0.4362037 0.3142593 2.730556 0.07007407
## free.sulfur.dioxide total.sulfur.dioxide density pH sulphates alcohol
## 1 15.97770 48.73826 0.996787 3.312230 0.6499296 10.32113
## 2 17.44444 43.70370 0.995672 3.346111 0.7405556 11.63642
## quality quality.label
## 1 5.464789 1
## 2 7.055556 2

#7.Variable importance: Random Forest can tell us which variable (features) are
#the most important for making predictions

wine.rf <- randomForest(quality.label ~.-quality, #all features except quality

data = wine.df.train, ntree = 3000,
importance = TRUE)
wine.rf

##
## Call:
## randomForest(formula = quality.label ~ . - quality, data = wine.df.train, ntree = 3000, importa
## Type of random forest: classification
## Number of trees: 3000
## No. of variables tried at each split: 3
##
## OOB estimate of error rate: 9.12%
## Confusion matrix:
## Bad Good class.error
## Bad 927 29 0.03033473
## Good 73 90 0.44785276

13
importance(wine.rf)

## Bad Good MeanDecreaseAccuracy MeanDecreaseGini

## fixed.acidity 23.07886 43.93875 45.88104 20.57973
## volatile.acidity 21.12378 83.23243 68.21415 33.76012
## citric.acid 13.31114 49.27367 48.04063 27.22622
## residual.sugar 20.44845 34.10328 36.78252 15.90856
## chlorides 27.10404 33.37016 40.62654 18.56683
## free.sulfur.dioxide 33.30032 38.04191 48.07403 16.90554
## total.sulfur.dioxide 32.11146 65.17008 59.14898 23.70860
## density 38.37314 46.22053 57.45747 25.28253
## pH 16.59061 42.56024 38.01753 16.75284
## sulphates 24.41116 93.79164 78.67998 33.30075
## alcohol 42.02788 99.17418 91.10301 46.77968

#---Variable importance plot

varImpPlot(wine.rf, main = "Feature Importance for Wine Quality Prediction")

Feature Importance for Wine Quality Prediction

alcohol alcohol
sulphates volatile.acidity
volatile.acidity sulphates
total.sulfur.dioxide citric.acid
density density
free.sulfur.dioxide total.sulfur.dioxide
citric.acid fixed.acidity
fixed.acidity chlorides
chlorides free.sulfur.dioxide
pH pH
residual.sugar residual.sugar

40 60 80 0 20 40
MeanDecreaseAccuracy MeanDecreaseGini

#---Heatmap for variable important

library(gplots)

## Warning: package ’gplots’ was built under R version 4.3.3

##
## Attaching package: ’gplots’

14
## The following object is masked from ’package:stats’:
##
## lowess

library(RColorBrewer)
heatmap.2(t(importance(wine.rf)[,1:2]), key = FALSE, col = brewer.pal(9, "Blues"),
dend = "none", trace = "none", margins = c(10,10),
main = "\n\n\nHeatmap of Feature Importance \n for Wine Quality Prediction",
cexRow = 1.5)

Heatmap of Feature Importance

for Wine Quality Prediction

Good

Bad
alcohol
volatile.acidity
sulphates
total.sulfur.dioxide
free.sulfur.dioxide
density
residual.sugar
chlorides
citric.acid
pH
fixed.acidity

15