Furhter information and code to run the experiment from ``A Constrained Cluster Ensemble Using Hierarchical Clustering Methods'' 1
Run RunExperiment.m in the Code directory for the main experiment (or RunEnsemble.m for the ensemble study).
Important
The Statistics and Machine Learning Toolbox is required to run this code.
The Data files are structured as such: <[features], class label>, where the row number is also the object number. These can be found in the Versions directory.
The Constraint files are in the format: <object 1, object 2, constraint type>, where constraint type is 1 or 0 depending on whether the object 1 and object 2 belong in the same cluster or must not be in the same cluster respectively. These can be found in the Constrains directory.
Note
The file structure in the directories is explained in 1
Detailed below are the attributes of the datasets from both the Real and Synthetic collections.
| Dataset | No. of Data Points | No. of Features | No. of Clusters |
|---|---|---|---|
| abalone | 4177 | 8 | 3 |
| acute-inflammation | 120 | 6 | 2 |
| acute-nephritis | 120 | 6 | 2 |
| adult | 48842 | 14 | 2 |
| annealing | 850 | 31 | 4 |
| arrhythmia | 295 | 262 | 10 |
| balance-scale | 576 | 4 | 3 |
| bank | 4521 | 16 | 2 |
| blood | 748 | 4 | 2 |
| breast-cancer-wisc-diag | 569 | 30 | 2 |
| breast-cancer-wisc | 699 | 9 | 2 |
| breast-cancer | 286 | 9 | 2 |
| car | 1728 | 6 | 4 |
| cardiotocography-10clases | 2126 | 21 | 10 |
| cardiotocography-3clases | 2126 | 21 | 3 |
| chess-krvk | 28029 | 6 | 18 |
| chess-krvkp | 3196 | 36 | 2 |
| congressional-voting | 435 | 16 | 2 |
| conn-bench-sonar-mines-rocks | 208 | 60 | 2 |
| conn-bench-vowel-deterding | 990 | 11 | 11 |
| connect-4 | 67557 | 42 | 2 |
| contrac | 1473 | 9 | 3 |
| credit-approval | 690 | 15 | 2 |
| cylinder-bands | 512 | 35 | 2 |
| dermatology | 297 | 34 | 5 |
| ecoli | 272 | 7 | 3 |
| energy-y1 | 768 | 8 | 3 |
| energy-y2 | 768 | 8 | 3 |
| glass | 146 | 9 | 2 |
| haberman-survival | 306 | 3 | 2 |
| hayes-roth | 129 | 3 | 2 |
| heart-cleveland | 219 | 13 | 2 |
| heart-hungarian | 294 | 12 | 2 |
| heart-va | 107 | 12 | 2 |
| hill-valley | 1212 | 100 | 2 |
| horse-colic | 368 | 25 | 2 |
| ilpd-indian-liver | 583 | 9 | 2 |
| image-segmentation | 2310 | 18 | 7 |
| ionosphere | 351 | 33 | 2 |
| iris | 150 | 4 | 3 |
| led-display | 1000 | 7 | 10 |
| letter | 20000 | 16 | 26 |
| low-res-spect | 469 | 100 | 4 |
| lymphography | 142 | 18 | 3 |
| magic | 19020 | 10 | 2 |
| mammographic | 961 | 5 | 2 |
| molec-biol-promoter | 106 | 57 | 2 |
| molec-biol-splice | 3190 | 60 | 3 |
| monks-1 | 556 | 6 | 2 |
| monks-2 | 601 | 6 | 2 |
| monks-3 | 554 | 6 | 2 |
| mushroom | 8124 | 21 | 2 |
| musk-1 | 476 | 166 | 2 |
| musk-2 | 6598 | 166 | 2 |
| nursery | 12958 | 8 | 5 |
| oocytes_merluccius_nucleus_4d | 1022 | 41 | 2 |
| oocytes_merluccius_states_2f | 1022 | 25 | 3 |
| oocytes_trisopterus_nucleus_2f | 912 | 25 | 2 |
| oocytes_trisopterus_states_5b | 898 | 32 | 3 |
| optical | 5620 | 62 | 10 |
| ozone | 2536 | 72 | 2 |
| page-blocks | 5445 | 10 | 5 |
| pendigits | 10992 | 16 | 10 |
| pima | 768 | 8 | 2 |
| planning | 182 | 12 | 2 |
| ringnorm | 7400 | 20 | 2 |
| seeds | 210 | 7 | 3 |
| semeion | 1593 | 256 | 10 |
| soybean | 362 | 35 | 15 |
| spambase | 4601 | 57 | 2 |
| spect | 265 | 22 | 2 |
| spectf | 267 | 44 | 2 |
| statlog-australian-credit | 690 | 14 | 2 |
| statlog-german-credit | 1000 | 24 | 2 |
| statlog-heart | 270 | 13 | 2 |
| statlog-image | 2310 | 18 | 7 |
| statlog-landsat | 6435 | 36 | 6 |
| statlog-shuttle | 57977 | 9 | 5 |
| statlog-vehicle | 846 | 18 | 4 |
| steel-plates | 1941 | 27 | 7 |
| synthetic-control | 600 | 60 | 6 |
| teaching | 102 | 5 | 3 |
| thyroid | 7200 | 21 | 3 |
| tic-tac-toe | 958 | 9 | 2 |
| titanic | 2201 | 3 | 2 |
| twonorm | 7400 | 20 | 2 |
| vertebral-column-2clases | 310 | 6 | 2 |
| vertebral-column-3clases | 310 | 6 | 3 |
| wall-following | 5456 | 24 | 4 |
| waveform-noise | 5000 | 40 | 3 |
| waveform | 5000 | 21 | 3 |
| wine-quality-red | 1571 | 11 | 5 |
| wine-quality-white | 4873 | 11 | 6 |
| wine | 130 | 13 | 2 |
| yeast | 1350 | 8 | 5 |
| Dataset | No. of Data Points | No. of Features | No. of Clusters |
|---|---|---|---|
| aggregation | 788 | 2 | 7 |
| aligned_bananas | 500 | 2 | 2 |
| arcs | 104 | 2 | 4 |
| balls_and_baguettes | 500 | 2 | 5 |
| bars | 500 | 2 | 2 |
| boat | 500 | 2 | 3 |
| cigar | 500 | 2 | 4 |
| circle_2_rectangles | 500 | 2 | 3 |
| circle_and_3_gaussians | 500 | 2 | 4 |
| concentric_circles_3 | 500 | 2 | 3 |
| enclosure | 622 | 2 | 3 |
| filled_circle | 500 | 2 | 2 |
| filled_circle_2 | 500 | 2 | 4 |
| flower | 500 | 2 | 5 |
| four_corners_clear | 500 | 2 | 4 |
| four_corners_noise | 500 | 2 | 4 |
| four_lines | 500 | 2 | 4 |
| gaussians_1_big_2_small | 500 | 2 | 3 |
| gaussians_3_touching | 500 | 2 | 3 |
| gaussians_5_compact | 500 | 2 | 5 |
| gaussians_5_unequal | 500 | 2 | 5 |
| gestalt | 399 | 2 | 6 |
| half_rings | 500 | 2 | 2 |
| happy_wave | 500 | 2 | 2 |
| orange | 500 | 2 | 2 |
| petals | 500 | 2 | 4 |
| random1 | 500 | 2 | 4 |
| random2 | 500 | 2 | 6 |
| random3 | 500 | 2 | 7 |
| randomised_normal | 500 | 2 | 9 |
| randomised_triangle | 500 | 2 | 10 |
| saturn | 500 | 2 | 2 |
| sixteen_blocks | 256 | 2 | 16 |
| spirals | 500 | 2 | 3 |
| stormclouds | 500 | 2 | 2 |
| t_and_u | 500 | 2 | 2 |
| ten_spherical | 500 | 2 | 10 |
| three_by_three | 500 | 2 | 9 |
| three_circles | 500 | 2 | 3 |
| two_diamonds | 500 | 2 | 2 |
| two_u | 260 | 2 | 2 |
| wingnut | 500 | 2 | 2 |
| worms | 500 | 2 | 4 |
| xor | 500 | 2 | 4 |
| xor_big_and_small | 500 | 2 | 4 |
| xor_different_cardinalities | 500 | 2 | 4 |
| yin_yang | 515 | 2 | 4 |
| atom | 500 | 3 | 2 |
| chainlink | 500 | 3 | 2 |
| hepta | 500 | 3 | 7 |
| tetra | 500 | 3 | 4 |
| torus_and_rod | 500 | 3 | 2 |