StreamSpot Bootstrap Clusters

00-StreamSpot-Bootstrap-Clusters.md

StreamSpot Bootstrap Clusters

www3.cs.stonybrook.edu/~emanzoor/streamspot/

Below are the bootstrap clusters used for the experiments in the StreamSpot paper for each of following datasets:

• all (01-C50_k10_all.txt): Chunk length of 50, 10 clusters.
• ydc (02-C25_k5_ydc.txt): Chunk length of 25, 5 clusters.
• gfc (03-C50_k5_gfc.txt): Chunk length of 50, 5 clusters.

The bootstrap clusters were generated as follows:

1. Sample 75% of the respective dataset.
2. Select a chunk length as specified in the paper.
3. Select the number of clusters K by running 10 trials of K-medoids and picking K with the maximum silhoutte coefficient on the resulting clustering.
4. Set the anomaly threshold for each cluster as 3 standard deviations away from the mean graph-to-medoid distance of that cluster.
5. Set the global threshold as 3 standard deviations away from the mean graph-to-medoid distance of all clusters.

The format of each bootstrap clusters file is as follows (all numbers are TAB separated):

number_of_clusters global_threshold
cluster_threshold_1 cluster_1_graph_id_1 cluster_1_graph_id_2 ...
cluster_threshold_2 cluster_2_graph_id_1 cluster_2_graph_id_2 ...
...

For any questions, please contact:

• [email protected]
• [email protected]

01-C50_k10_all.txt

10	0.4823
0.4341	80	79	25	15	39	40	53	17	57	50	18	69	87	16	47	3	38	52	8	34	44	72	59	91	98	14	21	12	58	82	95	86	76	54	90	42	32	23	37	62	9	1	45	75	55	81	92	99	36	56	13	46	27	24	28	65	7	88	61	97	77	73	63	29	0	51	10	74	67	66	60	84	85	30	89	115	273	278	277	213	224	280	286	211	237	227	272	229	292	268	258	285	206	209	298	261	282	216	251	207	200	270	256	239	234	230	263	294	220	248	284	244	228	293	217	214	208	281	210	225	297	291	205	202	222	231	218	249	215	241	265	295	204	232	243	279	276	274	254	266	233	287	219	221	253	212	246	264	203	235	283	542	535	552
0.0300	465	473	498	452	479	466	437	486	462	476	467	497	496	472	566	507	508	527
0.7231	150	173	120	169	110	125	191	187	128	189	164	132	183	134	197	119	151	144	163	180	179	171	140	102	185	113	104	126	155	116	158	176	196	157	174	162	114	133	455	416
0.1182	160	112	145	105	108	139	188	181	131	199	123	182	124	166	193	175	129	154	186	168	138	184	137	101	143	152	148	149	161	147	109	136	106	194	177	130
0.0014	442	491	434	485	412	492	420	448	495	463	407	422	429	402	411	417	431	470	449	428	421	406	446	409	458	460	403	435	440	401	419	487	484	405	499	418	444	477	408	461	469	427	424	413
0.1967	505	530	588	578	591	579	555	574	514	502	547	548	506	519	524	550	531
0.0041	438	464	468	447	430	439	475	423	459	450
0.4854	509	569	595	573	583	523	510	517	541	544	533	526	554	558	571	534	537	511	584	540	587	593	594	585	515	560	522	543	516	568	572	546	559	599	556	582	538	561	504	549	525
0.0046	489	481	482	493	443
0.2012	580	521	518	501	590	567	596	589	539	529

02-C25_k5_ydc.txt

5 0.9742
0.3526  79  75  63  15  38  11  40  45  39  86  4 47  27  81  17  33  9 92  72  69  67  2
6  68  77  80  20  32  51  0 54  10  36  21  48  6 61  90  25  24  93  95  85  41  99  97  1 16  5 64  18  89  34  42  35  43  57  23  46  44  60  78  94  96  13  98  3 22  552
0.9692  512 573 550 537 506 509 595 598 571 568 592 560 526 544 504 554 591 524 502 559 535 519 542 572 511 503 546 549 599 584 538 588 540 531 541 548 561 593 583 533 556 570 577 523 530 515 576 547 553 575 516 555 514 578 558 543
1.1399  459 405 483 435 408 474 445 412 436 428 414 427 453 487 482 485 461 438 429 424 479 486 488 469 419 496 421 448 431 443 480 411 472 475 466 491 437 473 467 462 465 471 456 430 447 423 425 401 457 450 441 484 476 477 498 446 434 481 464 439 400 454 449 433 451 499 413 418 458 494 440 478 409 402 489 528
0.4884  520 567 527 501 586 518 513 521 590 596 539 507 529 565 589 580 566
0.5650  62  82  66  2 76  29  28  87

03-C50_k10_gfc.txt

10  1.0288
1.0840  550 506 586 592 591 524 502 519 588 531 548 577 530 576 547 555 514 578
1.2903  286 205 257 208 273 212 235 214 263 281 237 228 223 278 287 282 220 230 242 274 2
65 254 262 232 270 255 246 222 224 201 269 249 283 267 245 233 260 200 213 217
1.1358  115 135 236 272 252 264 238 253
0.3811  145 139 186 147 181 182 109 166 167 168 154 136 148 106 129 193 195 143
0.6771  179 163 140 162 104 117 133 192 172 169 126 102 176 180 120 132 151 110 190 125 185 197 116 164 118 189 134 157 146 144 178 196 113 128 187 198 103
0.5603  528 520 567 527 501 518 513 521 590 596 539 507 529 565 589 580 566
1.0073  244 226 290 218 296 279 211 271 276 229 275 298 280 248 250 299 209 542
1.1986  234 227 259 284 285 247 240 256 294 239 277 202 535 552
1.0617  512 573 537 509 595 598 571 568 560 526 544 504 554 559 572 511 503 546 549 599 584 538 540 541 561 593 583 533 556 570 523 515 553 575 516 558 543
0.6206  175 138 111 127 177 100 121 161 124 141 199 101 105 142 123 160 194 122