File size: 46,667 Bytes
0d00d62 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 |
{
"title": "Transductive SVM Mastery: 100 MCQs",
"description": "A structured 3-level mastery set of 100 MCQs on Transductive Support Vector Machines (TSVM) — covering fundamentals, intuition, semi-supervised learning strategy, margin optimization, unlabeled data influence, and real-world scenario-based problem-solving.",
"questions": [
{
"id": 1,
"questionText": "What type of learning does Transductive SVM belong to?",
"options": [
"Unsupervised learning",
"Reinforcement learning",
"Supervised learning",
"Semi-supervised learning"
],
"correctAnswerIndex": 3,
"explanation": "TSVM uses both labeled and unlabeled data, making it a semi-supervised learning method."
},
{
"id": 2,
"questionText": "What is the main goal of Transductive SVM?",
"options": [
"Label only the given test dataset",
"Reduce number of features",
"Perform clustering",
"Predict all future data points"
],
"correctAnswerIndex": 0,
"explanation": "TSVM directly optimizes predictions for the given test set instead of generalizing globally."
},
{
"id": 3,
"questionText": "Which type of data does TSVM use during training?",
"options": [
"No data at all",
"Both labeled and unlabeled data",
"Only labeled data",
"Only unlabeled data"
],
"correctAnswerIndex": 1,
"explanation": "TSVM is semi-supervised and uses both labeled and unlabeled data."
},
{
"id": 4,
"questionText": "TSVM is mainly used to improve performance when labeled data is:",
"options": [
"Balanced",
"Abundant",
"Noisy",
"Very limited"
],
"correctAnswerIndex": 3,
"explanation": "TSVM performs best when labeled data is scarce but unlabeled data is available."
},
{
"id": 5,
"questionText": "What does TSVM try to optimize in its classification boundary?",
"options": [
"Random margin",
"Minimum number of support vectors",
"Margin using both labeled and unlabeled data",
"Widest margin between clusters"
],
"correctAnswerIndex": 2,
"explanation": "TSVM adjusts the margin using both labeled and unlabeled samples."
},
{
"id": 6,
"questionText": "Transductive SVM is different from traditional SVM because it:",
"options": [
"Uses gradient descent",
"Uses kernels",
"Works only on images",
"Uses unlabeled test data during training"
],
"correctAnswerIndex": 3,
"explanation": "TSVM includes test (unlabeled) data during training."
},
{
"id": 7,
"questionText": "Why does TSVM consider unlabeled data?",
"options": [
"To balance datasets",
"To improve generalization for test data",
"To reduce memory usage",
"To remove irrelevant features"
],
"correctAnswerIndex": 1,
"explanation": "Unlabeled data helps TSVM better align boundaries for the actual test samples."
},
{
"id": 8,
"questionText": "TSVM belongs to the category of:",
"options": [
"Fully supervised learning",
"Active learning",
"Semi-supervised learning",
"Reinforcement learning"
],
"correctAnswerIndex": 2,
"explanation": "TSVM is a semi-supervised learning method."
},
{
"id": 9,
"questionText": "What is the typical limitation of TSVM?",
"options": [
"Works only offline",
"Requires millions of labeled samples",
"Cannot handle text data",
"Computationally expensive"
],
"correctAnswerIndex": 3,
"explanation": "TSVM is more computationally heavy because of unlabeled data optimization."
},
{
"id": 10,
"questionText": "Transductive learning focuses on:",
"options": [
"Future unseen data",
"Only currently given test samples",
"Generating synthetic data",
"Creating new labels"
],
"correctAnswerIndex": 1,
"explanation": "Transductive learning makes predictions only for the given test samples."
},
{
"id": 11,
"questionText": "Which assumption helps TSVM improve classification?",
"options": [
"Data is purely random",
"Labels are dynamic",
"Unlabeled data follows the same distribution as labeled data",
"Classes always overlap"
],
"correctAnswerIndex": 2,
"explanation": "TSVM assumes unlabeled data follows the same distribution as labeled data."
},
{
"id": 12,
"questionText": "In TSVM, unlabeled data helps mainly in:",
"options": [
"Balancing training batches",
"Data compression",
"Changing learning rate",
"Shifting decision boundary to better margin"
],
"correctAnswerIndex": 3,
"explanation": "Unlabeled data helps TSVM adjust the decision boundary more accurately."
},
{
"id": 13,
"questionText": "TSVM is best suited when unlabeled data is:",
"options": [
"Much more than labeled data",
"Less than labeled data",
"Exactly equal to labeled data",
"Not available at all"
],
"correctAnswerIndex": 0,
"explanation": "TSVM benefits most when unlabeled data is available in large quantity."
},
{
"id": 14,
"questionText": "TSVM mainly helps solve which challenge?",
"options": [
"Memory usage",
"Feature selection",
"Label scarcity",
"Overfitting"
],
"correctAnswerIndex": 2,
"explanation": "TSVM improves performance even with very few labeled samples."
},
{
"id": 15,
"questionText": "TSVM does not require:",
"options": [
"Margin calculation",
"Labeled test data",
"Kernel function",
"Optimization"
],
"correctAnswerIndex": 1,
"explanation": "TSVM requires unlabeled test data, not labeled test data."
},
{
"id": 16,
"questionText": "TSVM selects decision boundaries by:",
"options": [
"Minimizing random noise",
"Maximizing margin with unlabeled support",
"Maximizing entropy",
"Ignoring unlabeled samples"
],
"correctAnswerIndex": 1,
"explanation": "TSVM pushes boundary into low-density unlabeled regions."
},
{
"id": 17,
"questionText": "Transductive SVM improves performance mainly by:",
"options": [
"Gradient clipping",
"Feature reduction",
"Boundary adjustment using unlabeled data",
"Random sampling"
],
"correctAnswerIndex": 2,
"explanation": "Unlabeled data allows TSVM to refine classification boundary."
},
{
"id": 18,
"questionText": "Which of the following is true for TSVM?",
"options": [
"Requires labeled test data",
"Uses test data during training",
"Uses only training data",
"Does not use kernel trick"
],
"correctAnswerIndex": 1,
"explanation": "TSVM includes test data in the optimization loop."
},
{
"id": 19,
"questionText": "Which situation best suits TSVM?",
"options": [
"Only structured data is present",
"No test data is known",
"Few labeled but many unlabeled samples exist",
"Lots of labeled data"
],
"correctAnswerIndex": 2,
"explanation": "TSVM is ideal when unlabeled data is abundant but labels are limited."
},
{
"id": 20,
"questionText": "TSVM differs from supervised SVM mainly because it:",
"options": [
"Uses unlabeled data during optimization",
"Uses learning rate",
"Does not optimize margin",
"Can work without data"
],
"correctAnswerIndex": 0,
"explanation": "Unlabeled data is a key differentiator in TSVM’s boundary optimization."
},
{
"id": 21,
"questionText": "TSVM attempts to place the decision boundary in:",
"options": [
"Random regions",
"Low-density regions",
"Fixed center of data",
"High-density regions"
],
"correctAnswerIndex": 1,
"explanation": "Best decision boundaries avoid dense regions of unlabeled data."
},
{
"id": 22,
"questionText": "Which term is most associated with TSVM?",
"options": [
"Auto encoding",
"Generalization only",
"Label propagation",
"Transductive inference"
],
"correctAnswerIndex": 3,
"explanation": "TSVM is a transductive inference-based model."
},
{
"id": 23,
"questionText": "Unlabeled test data in TSVM is used for:",
"options": [
"Feature scaling",
"Random dropout",
"Adjusting classification boundary",
"Validation only"
],
"correctAnswerIndex": 2,
"explanation": "TSVM modifies its boundary based on test samples’ structure."
},
{
"id": 24,
"questionText": "Transductive SVM is most helpful when:",
"options": [
"Labels change dynamically",
"No test data exists",
"Test set is provided beforehand",
"Only regression is required"
],
"correctAnswerIndex": 2,
"explanation": "TSVM needs test samples available during training."
},
{
"id": 25,
"questionText": "TSVM primarily tries to:",
"options": [
"Remove irrelevant classes",
"Cluster unlabeled samples",
"Predict all possible future distributions",
"Optimize boundary for specific test data"
],
"correctAnswerIndex": 3,
"explanation": "TSVM is optimized for performance on specific given test data."
},
{
"id": 26,
"questionText": "In TSVM, unlabeled points influence the model by:",
"options": [
"Being ignored during margin calculation",
"Randomly deciding class labels",
"Reducing kernel complexity",
"Helping move the boundary into low-density regions"
],
"correctAnswerIndex": 3,
"explanation": "Unlabeled points guide the boundary away from dense data areas to achieve better separation."
},
{
"id": 27,
"questionText": "What is the main optimization challenge in TSVM?",
"options": [
"It is a non-convex optimization problem",
"It ignores constraints completely",
"It minimizes feature size",
"Convex optimization is guaranteed"
],
"correctAnswerIndex": 0,
"explanation": "TSVM’s optimization is non-convex because unlabeled data labels are unknown and must be inferred during training."
},
{
"id": 28,
"questionText": "Which term best describes how TSVM assigns labels to unlabeled data during training?",
"options": [
"Rule-based labeling",
"Hard clustering",
"Automatic label inference",
"Supervised annotation"
],
"correctAnswerIndex": 2,
"explanation": "TSVM infers pseudo-labels for unlabeled data as part of its optimization process."
},
{
"id": 29,
"questionText": "What does TSVM attempt to minimize?",
"options": [
"Sum of absolute residuals",
"Classification error on both labeled and unlabeled data",
"Only training loss",
"Kernel bias"
],
"correctAnswerIndex": 1,
"explanation": "TSVM minimizes overall classification error including pseudo-labeled samples."
},
{
"id": 30,
"questionText": "The optimization objective of TSVM includes:",
"options": [
"Both labeled loss and unlabeled margin penalty",
"Only kernel regularization",
"Only labeled data loss",
"Clustering objective"
],
"correctAnswerIndex": 0,
"explanation": "TSVM jointly optimizes labeled loss and a penalty term for unlabeled margin consistency."
},
{
"id": 31,
"questionText": "How does TSVM improve over SVM when labeled data is scarce?",
"options": [
"By ignoring unlabeled samples",
"By using fixed weight initialization",
"By forcing boundary to pass through dense unlabeled data",
"By using unlabeled data to refine margin placement"
],
"correctAnswerIndex": 3,
"explanation": "Unlabeled samples help TSVM adjust its boundary even with limited labeled examples."
},
{
"id": 32,
"questionText": "Which assumption is most crucial for TSVM to work effectively?",
"options": [
"All classes have equal size",
"Unlabeled data is random noise",
"Labeled and unlabeled data come from similar distributions",
"All features are binary"
],
"correctAnswerIndex": 2,
"explanation": "TSVM assumes labeled and unlabeled data share the same underlying distribution."
},
{
"id": 33,
"questionText": "What is a pseudo-label in TSVM?",
"options": [
"An estimated label for unlabeled data",
"A feature scaling term",
"A random class label",
"A type of kernel parameter"
],
"correctAnswerIndex": 0,
"explanation": "TSVM assigns temporary labels (pseudo-labels) to unlabeled data during optimization."
},
{
"id": 34,
"questionText": "Why is TSVM computationally more expensive than SVM?",
"options": [
"It uses both labeled and unlabeled samples in optimization",
"It ignores regularization",
"It skips margin maximization",
"It trains multiple neural networks internally"
],
"correctAnswerIndex": 0,
"explanation": "Involving unlabeled samples increases optimization complexity in TSVM."
},
{
"id": 35,
"questionText": "In TSVM, pseudo-label assignment affects:",
"options": [
"Hyperparameter tuning only",
"Feature normalization",
"Decision boundary and margin",
"Kernel shape only"
],
"correctAnswerIndex": 2,
"explanation": "Pseudo-labels change how unlabeled data influences the decision boundary."
},
{
"id": 36,
"questionText": "TSVM is also known as:",
"options": [
"Support Vector Data Description",
"Hybrid Neural Classifier",
"Maximum Margin Estimator",
"Semi-supervised SVM"
],
"correctAnswerIndex": 3,
"explanation": "TSVM is often referred to as Semi-supervised SVM because it uses unlabeled data."
},
{
"id": 37,
"questionText": "The optimization of TSVM is often solved by:",
"options": [
"Greedy pruning of features",
"Alternating between label estimation and margin optimization",
"Using only test accuracy",
"Random initialization of labels and stopping early"
],
"correctAnswerIndex": 1,
"explanation": "TSVM alternates between assigning pseudo-labels and optimizing the decision boundary."
},
{
"id": 38,
"questionText": "TSVM tries to ensure that unlabeled points:",
"options": [
"Are misclassified intentionally",
"Are ignored completely",
"Fall near the decision boundary",
"Stay in low-density regions away from the boundary"
],
"correctAnswerIndex": 3,
"explanation": "TSVM prefers placing the boundary where few unlabeled points exist (low-density assumption)."
},
{
"id": 39,
"questionText": "What is the purpose of using test data in TSVM training?",
"options": [
"To tune kernel parameters",
"To calculate validation loss",
"To influence boundary for that specific test set",
"To estimate gradient descent steps"
],
"correctAnswerIndex": 2,
"explanation": "TSVM includes test data to directly optimize for that specific evaluation set."
},
{
"id": 40,
"questionText": "Which of the following is NOT an advantage of TSVM?",
"options": [
"Simple convex optimization",
"Uses unlabeled data effectively",
"Improved test-specific accuracy",
"Better performance with limited labeled data"
],
"correctAnswerIndex": 0,
"explanation": "TSVM’s optimization is non-convex, which can make it difficult to solve efficiently."
},
{
"id": 41,
"questionText": "What happens if unlabeled data comes from a different distribution in TSVM?",
"options": [
"It has no effect",
"Training loss becomes zero",
"Boundary may become misleading",
"Model accuracy improves"
],
"correctAnswerIndex": 2,
"explanation": "If unlabeled data doesn’t match labeled distribution, TSVM’s decision boundary may shift incorrectly."
},
{
"id": 42,
"questionText": "Which kernel can be used in TSVM?",
"options": [
"Any kernel supported by SVM",
"Only linear kernel",
"No kernel at all",
"Only polynomial kernel"
],
"correctAnswerIndex": 0,
"explanation": "TSVM supports any standard SVM kernel such as linear, RBF, or polynomial."
},
{
"id": 43,
"questionText": "The low-density separation principle in TSVM ensures:",
"options": [
"Boundary avoids dense regions of data",
"Model ignores test data",
"Optimization is convex",
"Boundary passes through dense data points"
],
"correctAnswerIndex": 0,
"explanation": "TSVM seeks decision boundaries in low-density regions for better generalization."
},
{
"id": 44,
"questionText": "Which is a potential drawback of TSVM?",
"options": [
"Ignores margin constraint",
"Cannot handle kernels",
"Overfitting on unlabeled data structure",
"Too few optimization parameters"
],
"correctAnswerIndex": 2,
"explanation": "TSVM may overfit to the unlabeled data structure if the assumption about data distribution fails."
},
{
"id": 45,
"questionText": "In TSVM optimization, the cost function includes terms for:",
"options": [
"Feature reduction only",
"Test accuracy and dropout",
"Labeled and unlabeled sample penalties",
"Training error and kernel bias"
],
"correctAnswerIndex": 2,
"explanation": "Both labeled and unlabeled sample penalties are part of TSVM’s cost function."
},
{
"id": 46,
"questionText": "The transductive setting assumes access to:",
"options": [
"Future unseen data points",
"Synthetic datasets only",
"Only current test dataset during training",
"No data at training time"
],
"correctAnswerIndex": 2,
"explanation": "TSVM is transductive because it knows the test data in advance."
},
{
"id": 47,
"questionText": "In TSVM, which samples become support vectors?",
"options": [
"Only labeled samples",
"Randomly chosen samples",
"Only unlabeled samples",
"Both labeled and unlabeled samples that define margin"
],
"correctAnswerIndex": 3,
"explanation": "Support vectors can include both labeled and unlabeled points that lie near the margin."
},
{
"id": 48,
"questionText": "What role does the kernel trick play in TSVM?",
"options": [
"Converts classification to clustering",
"Removes need for optimization",
"Reduces computation cost",
"Transforms data into higher dimensions for better separation"
],
"correctAnswerIndex": 3,
"explanation": "Kernels in TSVM allow nonlinear separation using higher-dimensional mappings."
},
{
"id": 49,
"questionText": "If the unlabeled data distribution overlaps both classes, TSVM may:",
"options": [
"Misplace the boundary",
"Achieve perfect separation",
"Find an ideal margin",
"Completely ignore labels"
],
"correctAnswerIndex": 0,
"explanation": "Overlapping distributions make boundary placement difficult, reducing TSVM accuracy."
},
{
"id": 50,
"questionText": "What is the optimization goal of TSVM?",
"options": [
"Minimize Euclidean distance",
"Maximize kernel variance",
"Minimize total hinge loss for labeled and pseudo-labeled data",
"Minimize feature weights"
],
"correctAnswerIndex": 2,
"explanation": "TSVM extends hinge loss to include pseudo-labeled unlabeled samples."
},
{
"id": 51,
"questionText": "The margin in TSVM is influenced by:",
"options": [
"Feature normalization",
"Unlabeled data placement",
"Random initialization",
"Only kernel parameters"
],
"correctAnswerIndex": 1,
"explanation": "Unlabeled data impacts how the margin is optimized in TSVM."
},
{
"id": 52,
"questionText": "TSVM can be sensitive to:",
"options": [
"Data sorting order",
"Kernel choice and unlabeled data quality",
"Training batch size only",
"Hardware configuration"
],
"correctAnswerIndex": 1,
"explanation": "Both kernel and unlabeled data quality strongly influence TSVM performance."
},
{
"id": 53,
"questionText": "The optimization of TSVM is often performed using:",
"options": [
"Backpropagation",
"Gradient descent only",
"Iterative label flipping and margin refinement",
"Reinforcement updates"
],
"correctAnswerIndex": 2,
"explanation": "TSVM optimization alternates between label flipping and boundary updates."
},
{
"id": 54,
"questionText": "The unlabeled samples in TSVM are assigned:",
"options": [
"Random labels once",
"Temporary pseudo-labels updated iteratively",
"No labels ever",
"Permanent labels"
],
"correctAnswerIndex": 1,
"explanation": "Pseudo-labels are iteratively refined as the model learns."
},
{
"id": 55,
"questionText": "What is the primary risk of incorrect pseudo-labels in TSVM?",
"options": [
"Reduced convergence speed",
"Lower feature correlation",
"Kernel instability",
"Boundary shift and wrong classification"
],
"correctAnswerIndex": 3,
"explanation": "Incorrect pseudo-labels can distort margin direction and harm performance."
},
{
"id": 56,
"questionText": "TSVM can be viewed as solving a problem that involves both:",
"options": [
"Clustering and reinforcement",
"Supervised and unsupervised objectives",
"Regression and clustering",
"Classification and regression trees"
],
"correctAnswerIndex": 1,
"explanation": "TSVM combines labeled (supervised) and unlabeled (unsupervised) objectives."
},
{
"id": 57,
"questionText": "TSVM uses the concept of margin maximization from:",
"options": [
"Decision trees",
"Neural networks",
"Traditional SVMs",
"Naive Bayes"
],
"correctAnswerIndex": 2,
"explanation": "Like SVM, TSVM also maximizes the decision margin."
},
{
"id": 58,
"questionText": "Which step in TSVM optimization determines pseudo-label accuracy?",
"options": [
"Initialization of weights",
"Label inference phase",
"Gradient adjustment",
"Kernel scaling"
],
"correctAnswerIndex": 1,
"explanation": "Label inference directly controls pseudo-label quality in TSVM."
},
{
"id": 59,
"questionText": "The iterative process in TSVM continues until:",
"options": [
"Training accuracy is 100%",
"Labels stop changing significantly",
"Random seed resets",
"Kernel converges to zero"
],
"correctAnswerIndex": 1,
"explanation": "TSVM stops when pseudo-labels stabilize between iterations."
},
{
"id": 60,
"questionText": "The optimization process in TSVM is considered:",
"options": [
"Convex and easy to solve",
"Always deterministic",
"Non-convex and computationally hard",
"Independent of initial labels"
],
"correctAnswerIndex": 2,
"explanation": "TSVM optimization is non-convex because of unlabeled data label uncertainty."
},
{
"id": 61,
"questionText": "What happens when unlabeled data contradicts labeled patterns in TSVM?",
"options": [
"Model ignores unlabeled data",
"Pseudo-labels become fixed",
"Decision boundary becomes unstable",
"Kernel stops updating"
],
"correctAnswerIndex": 2,
"explanation": "Contradictory unlabeled data can confuse boundary direction."
},
{
"id": 62,
"questionText": "Which of the following best describes TSVM’s loss function?",
"options": [
"Mean squared error",
"Exponential loss",
"Cross-entropy only",
"Standard hinge loss extended with unlabeled penalty terms"
],
"correctAnswerIndex": 3,
"explanation": "TSVM extends the hinge loss to include penalties for unlabeled pseudo-labels."
},
{
"id": 63,
"questionText": "The decision boundary in TSVM depends heavily on:",
"options": [
"Distribution of unlabeled samples",
"Number of epochs",
"Learning rate decay",
"Batch normalization"
],
"correctAnswerIndex": 0,
"explanation": "Unlabeled data distribution strongly influences margin placement in TSVM."
},
{
"id": 64,
"questionText": "Which element makes TSVM non-trivial to implement?",
"options": [
"Fixed kernel trick",
"No support vector calculation",
"Static dataset handling",
"Dynamic pseudo-label optimization"
],
"correctAnswerIndex": 3,
"explanation": "Dynamic pseudo-label optimization adds complexity to TSVM."
},
{
"id": 65,
"questionText": "Which kind of semi-supervised principle does TSVM follow?",
"options": [
"Cluster assumption",
"Low-density separation assumption",
"Graph-based propagation",
"Entropy minimization only"
],
"correctAnswerIndex": 1,
"explanation": "TSVM is based on low-density separation — boundaries avoid dense data regions."
},
{
"id": 66,
"questionText": "The decision function of TSVM is closest in structure to:",
"options": [
"Random forest vote",
"Standard SVM decision function",
"Neural classifier function",
"Logistic regression probability"
],
"correctAnswerIndex": 1,
"explanation": "TSVM shares the same functional form as SVM but with adjusted parameters from unlabeled data."
},
{
"id": 67,
"questionText": "TSVM optimization complexity grows with:",
"options": [
"Number of labels only",
"Number of unlabeled samples",
"Learning rate value",
"Feature normalization"
],
"correctAnswerIndex": 1,
"explanation": "Larger unlabeled datasets increase TSVM’s computational cost."
},
{
"id": 68,
"questionText": "In Transductive SVM, what is the primary goal when adjusting the decision boundary?",
"options": [
"To cluster data without any decision boundary",
"To maximize margin while correctly classifying both labeled and unlabeled data",
"To only minimize error on labeled data",
"To ignore the unlabeled data completely"
],
"correctAnswerIndex": 1,
"explanation": "Transductive SVM aims to find a boundary that maximizes the margin while making good use of both labeled and unlabeled data for improved generalization."
},
{
"id": 69,
"questionText": "What happens if unlabeled data points lie very close to the initial decision boundary in TSVM?",
"options": [
"The model forces them to labeled class randomly",
"TSVM ignores them completely",
"The boundary may be adjusted to push those points away from the margin",
"The boundary becomes fixed and cannot move"
],
"correctAnswerIndex": 2,
"explanation": "TSVM tries to adjust the boundary so unlabeled points do not lie close to or inside the margin region, improving generalization."
},
{
"id": 70,
"questionText": "Why is Transductive SVM considered more complex than Inductive SVM?",
"options": [
"It ignores labeled data",
"It never uses margin maximization",
"It optimizes label assignment for unlabeled data too",
"It is only used for unsupervised problems"
],
"correctAnswerIndex": 2,
"explanation": "TSVM adds an additional layer of optimization by trying all possible label assignments of unlabeled data while maximizing the margin."
},
{
"id": 71,
"questionText": "A company has 100 labeled reviews and 10,000 unlabeled reviews. Why might TSVM perform better than regular SVM here?",
"options": [
"Because it assumes all reviews are identical",
"Because it uses unlabeled reviews to better position the decision boundary",
"Because it randomly guesses the boundary",
"Because it fully ignores unlabeled reviews"
],
"correctAnswerIndex": 1,
"explanation": "TSVM benefits from large unlabeled data by adjusting the decision boundary based on its distribution."
},
{
"id": 72,
"questionText": "In a real-world email spam detection scenario, how does TSVM specifically help compared to standard SVM?",
"options": [
"It forces all unlabeled emails to spam class",
"It uses unlabeled emails to refine the margin location before finalizing decisions",
"It converts unlabeled emails to labeled by itself",
"It deletes unlabeled emails"
],
"correctAnswerIndex": 1,
"explanation": "TSVM leverages structure in unlabeled email data to position the margin more correctly before classification."
},
{
"id": 73,
"questionText": "If TSVM is applied to medical diagnosis where only 10% of data is labeled, what is its most valuable capability?",
"options": [
"Removing all uncertain data",
"Automatically labeling data without optimization",
"Ignoring labeled data",
"Shifting the boundary based on unlabeled data distribution"
],
"correctAnswerIndex": 3,
"explanation": "TSVM improves decision quality by adjusting boundary using the structure of unlabeled data."
},
{
"id": 74,
"questionText": "What kind of data scenario benefits MOST from TSVM?",
"options": [
"Very few labeled samples but large structured unlabeled data",
"Plenty of labeled samples and zero unlabeled samples",
"Fully noisy unlabeled data with no distribution pattern",
"No patterns in unlabeled data"
],
"correctAnswerIndex": 0,
"explanation": "TSVM is highly effective when labeled data is limited but unlabeled data is large and meaningful."
},
{
"id": 75,
"questionText": "Which real-world use case is a strong fit for Transductive SVM?",
"options": [
"Perfectly labeled tiny dataset with no unlabeled data",
"Pure clustering without labels",
"Handwritten text recognition with many unlabeled samples",
"Only regression problems"
],
"correctAnswerIndex": 2,
"explanation": "Handwriting recognition often has few labeled samples and many similar unlabeled data points, ideal for TSVM."
},
{
"id": 76,
"questionText": "During TSVM training, what happens if an unlabeled sample lies deep inside a wrong class region?",
"options": [
"Model stops training immediately",
"It forces all labeled points to change",
"TSVM pushes the margin so that it lies on the correct side if possible",
"It is permanently ignored"
],
"correctAnswerIndex": 2,
"explanation": "TSVM tries to shift the boundary gradually to avoid putting unlabeled samples deep into the wrong class."
},
{
"id": 77,
"questionText": "In TSVM, what is the usual effect of including many confidently separable unlabeled points?",
"options": [
"It makes decision boundary more stable and better aligned with data clusters",
"It makes model unstable and random",
"It causes TSVM to ignore all data",
"It forces decision boundary to flip constantly"
],
"correctAnswerIndex": 0,
"explanation": "Confidently separable unlabeled points reinforce stable decision boundary alignment with actual cluster structure."
},
{
"id": 78,
"questionText": "What type of challenge commonly occurs during TSVM optimization?",
"options": [
"It never converges by design",
"Non-convex optimization due to unknown unlabeled labels",
"No need for iteration",
"Linear time complexity"
],
"correctAnswerIndex": 1,
"explanation": "TSVM solves a non-convex optimization problem due to the additional label assignment search."
},
{
"id": 79,
"questionText": "Why is TSVM training usually slower than standard SVM?",
"options": [
"It ignores mathematics",
"It does not use margin maximization",
"It tries multiple label combinations for unlabeled data",
"It stops training halfway always"
],
"correctAnswerIndex": 2,
"explanation": "TSVM trains slower because it must attempt different label assignments while maximizing margin."
},
{
"id": 80,
"questionText": "In TSVM, what is the main purpose of using unlabeled data?",
"options": [
"To randomly flip decision boundaries",
"To confuse the model",
"To better shape the margin according to real data distribution",
"To delete labeled samples"
],
"correctAnswerIndex": 2,
"explanation": "Unlabeled data helps define a boundary that mirrors actual data cluster structure."
},
{
"id": 81,
"questionText": "In a fraud detection task, why can TSVM be extremely useful?",
"options": [
"Because fraud data is always labeled",
"Because it ignores real-world data distribution",
"Because TSVM works only with labeled data",
"Because majority of data is unlabeled and TSVM can learn from it"
],
"correctAnswerIndex": 3,
"explanation": "TSVM is powerful in fraud detection where labeled fraud cases are rare but transaction patterns (unlabeled) are massive."
},
{
"id": 82,
"questionText": "Which scenario would likely MISLEAD a TSVM model?",
"options": [
"Unlabeled data that is all random noise or mixed distributions",
"Large structured unlabeled data",
"High-margin clusterable data",
"Unlabeled data that follows clear separable clusters"
],
"correctAnswerIndex": 0,
"explanation": "If unlabeled data is highly noisy or structureless, TSVM may push margins incorrectly."
},
{
"id": 83,
"questionText": "What happens if all unlabeled data points strongly overlap between classes in TSVM?",
"options": [
"TSVM forces them into positive class",
"Model stops and refuses to run",
"TSVM may struggle to move the margin meaningfully",
"TSVM easily finds the perfect boundary"
],
"correctAnswerIndex": 2,
"explanation": "If unlabeled samples are non-separable, TSVM gets little advantage since it cannot adjust margin meaningfully."
},
{
"id": 84,
"questionText": "When applying TSVM to a dataset with label noise in the labeled data, what might happen?",
"options": [
"TSVM deletes the noisy samples",
"TSVM ignores unlabeled data",
"TSVM improves automatically with noise",
"TSVM may amplify wrong boundary decisions"
],
"correctAnswerIndex": 3,
"explanation": "If labeled data is noisy, TSVM might reinforce the wrong boundary because it follows a misleading starting label set."
},
{
"id": 85,
"questionText": "Which TSVM behavior is desirable in real world customer segmentation?",
"options": [
"Using no unlabeled input in training",
"Adjusting decision boundary using natural grouping of unlabeled customers",
"Deleting all unlabeled customers",
"Blindly guessing labels"
],
"correctAnswerIndex": 1,
"explanation": "TSVM benefits segmentation by aligning its boundary with naturally existing customer groups found in unlabeled data."
},
{
"id": 86,
"questionText": "In TSVM-based image classification, what advantage does it provide during learning?",
"options": [
"It forces all images into one class",
"It shapes the decision boundary using the natural distribution of unlabeled images",
"It uses only textual features",
"It refuses to learn if labels are missing"
],
"correctAnswerIndex": 1,
"explanation": "TSVM adapts boundary using unlabeled images' pattern distribution, improving classification accuracy."
},
{
"id": 87,
"questionText": "What is a potential risk if unlabeled data is from a DIFFERENT distribution than labeled data in TSVM?",
"options": [
"TSVM asks user to delete data",
"It has no effect on boundary",
"TSVM automatically fixes it",
"Performance may degrade because TSVM forces boundary based on misleading structure"
],
"correctAnswerIndex": 3,
"explanation": "If unlabeled data is from a different domain, TSVM may wrongly adjust boundary, hurting accuracy."
},
{
"id": 88,
"questionText": "A TSVM is trained on 5 labeled dog images and 2000 unlabeled animal images. What is it most likely to do?",
"options": [
"Stop training",
"Ignore the dog images",
"Use unlabeled animal patterns to identify the dog boundary more accurately",
"Randomly assign all unlabeled to dog"
],
"correctAnswerIndex": 2,
"explanation": "TSVM will exploit unlabeled animal images to shape a correct separation boundary for identifying dogs."
},
{
"id": 89,
"questionText": "How does TSVM generally react if unlabeled points are clearly clustered far from each other?",
"options": [
"It positions margin between clusters for better generalization",
"It fails to separate them",
"It merges clusters manually",
"It shuts down"
],
"correctAnswerIndex": 0,
"explanation": "TSVM naturally places its margin between unlabeled clusters if they are well separated."
},
{
"id": 90,
"questionText": "In TSVM, what outcome indicates that unlabeled data has genuinely improved learning?",
"options": [
"Lower margin size",
"Higher generalization accuracy on test data",
"Model predicts everything as same class",
"Model ignores patterns"
],
"correctAnswerIndex": 1,
"explanation": "Proper use of unlabeled data leads to improved generalization accuracy in TSVM."
},
{
"id": 91,
"questionText": "When might TSVM produce WORSE results than Inductive SVM?",
"options": [
"When unlabeled data is huge but structured",
"When unlabeled data comes from a completely different unrelated distribution",
"When labeled data is clean and unlabeled data is small",
"When unlabeled data perfectly follows same distribution"
],
"correctAnswerIndex": 1,
"explanation": "If unlabeled data is from a different domain, TSVM may adjust margin wrongly leading to worse performance."
},
{
"id": 92,
"questionText": "A TSVM model is being applied to sentiment analysis with mixed noisy unlabeled social media data. What must be carefully monitored?",
"options": [
"That it ignores unlabeled data completely",
"That all unlabeled data is forcibly labeled positive",
"That unlabeled data does not distort margin due to noise",
"That it deletes all neutral comments"
],
"correctAnswerIndex": 2,
"explanation": "Noisy unlabeled sentiment data may badly affect TSVM, so quality of unlabeled data must be monitored."
},
{
"id": 93,
"questionText": "What is the ideal unlabeled data characteristic for TSVM success?",
"options": [
"Highly overlapping with random noise",
"Totally unrelated to task",
"Completely identical to labeled data",
"Clearly structured with separable cluster tendencies"
],
"correctAnswerIndex": 3,
"explanation": "TSVM excels when unlabeled data is structured in distinct patterns that help form clear separation margins."
},
{
"id": 94,
"questionText": "In practice, what process is often done BEFORE using TSVM on real data?",
"options": [
"Converting unlabeled to random labels",
"Deleting all unlabeled data",
"Data cleaning and verifying unlabeled distribution quality",
"Removing margin maximization"
],
"correctAnswerIndex": 2,
"explanation": "Proper preprocessing is essential to ensure unlabeled data is reliable before applying TSVM."
},
{
"id": 95,
"questionText": "If unlabeled data heavily contradicts labeled examples in TSVM, what might occur?",
"options": [
"TSVM successfully corrects labels always",
"TSVM perfectly separates everything",
"TSVM automatically detects noise and stops",
"TSVM may get confused and produce poor boundary"
],
"correctAnswerIndex": 3,
"explanation": "Contradictory unlabeled data can mislead TSVM into placing wrong boundaries."
},
{
"id": 96,
"questionText": "In a multi-class extension of TSVM, what additional challenge appears?",
"options": [
"It becomes unsupervised",
"It stops supporting margin maximization",
"Complexity increases with relabeling across multiple decision boundaries",
"No unlabeled data is used"
],
"correctAnswerIndex": 2,
"explanation": "Multi-class TSVM must explore boundary adjustments across multiple possible label assignments, raising complexity."
},
{
"id": 97,
"questionText": "What strategy can help TSVM avoid overfitting wrongly to noisy unlabeled data?",
"options": [
"Weight control or confidence filtering on unlabeled samples",
"Completely ignore margin",
"Force all unlabeled data to positive class",
"Never use labeled data"
],
"correctAnswerIndex": 0,
"explanation": "Confidence-based control helps TSVM use only reliable unlabeled points for boundary adjustment."
},
{
"id": 98,
"questionText": "In TSVM-based product recommendation, what is the main benefit of leveraging unlabeled browsing data?",
"options": [
"To delete purchase history",
"To adjust the decision function according to real user interest patterns",
"To cluster users without any classification",
"To force all users into one preference group"
],
"correctAnswerIndex": 1,
"explanation": "Browsing patterns (unlabeled) help TSVM refine decision boundaries to better match actual user preferences."
},
{
"id": 99,
"questionText": "When deployed in semi-supervised security systems, what critical aspect must be monitored in TSVM?",
"options": [
"It must run without labeled data",
"It never needs retraining",
"Unlabeled threat data must be from reliable and relevant distribution",
"All unlabeled threats are ignored"
],
"correctAnswerIndex": 2,
"explanation": "TSVM relies on unlabeled data being relevant to security context; irrelevant sources can shift boundary incorrectly."
},
{
"id": 100,
"questionText": "What best represents a real-world success condition for TSVM?",
"options": [
"Unlabeled structured perfectly, labeled data small",
"Labeled data large, unlabeled zero",
"Unlabeled completely unrelated garbage",
"Labeled data randomly mislabeled"
],
"correctAnswerIndex": 0,
"explanation": "TSVM is powerful when unlabeled data is abundant and structured, while labeled data is limited."
}
]
}
|