PenDigits (version#06)

The 10 classes contained in this data set correspond to the digits from 0 to 9, with examples created by different hand writings. Class 4, defined here as outlier, was downsampled to only 20 objects. After the preprocessing, this database has 16 numeric attributes and 9868 instances, divided into 20 outliers (0.2%) and 9848 inliers (99.8%). This dataset is already normalized, i.e., all 16 attributes (spatial coordinates) have the same range [0,100]. It has been used in this form in [1,2].

References:

[1] H.-P. Kriegel, P. Kroeger, E. Schubert, and A. Zimek. Interpreting and unifying outlier scores. In Proc. SDM, pages 13-24, 2011.
[2] E. Schubert, R. Wojdanowski, A. Zimek, and H.-P. Kriegel. On evaluation of outlier rankings and outlier scores. In Proc. SDM, pages 1047-1058, 2012.

Download all data set variants used (2.1 MB). You can also access the original data. (merge train and test [pendigits.tes and pendigits.tra])

Normalized, without duplicates

This version contains 16 attributes, 9868 objects, 20 outliers (0.20%)

Download raw algorithm results (83.1 MB) Download raw algorithm evaluation table (56.3 kB)

Best Parameters

The following table contains the best (overall and per-method) results for each method and evaluation measure (when the same score was achieved twice, only the smallest k is given).
The Maximum F1-Measure is complimentary in addition to the measures in the original publication.

Algorithm	k	P@n	Adj. P@n	AP	Adj. AP	Max-F1	Adj. MF1	ROC AUC
KNN	5	0.05000	0.04807	0.07238	0.07050	0.15714	0.15543	0.98396
KNN	11	0.00000	-0.00203	0.09878	0.09695	0.20635	0.20474	0.98912
KNN	12	0.00000	-0.00203	0.09931	0.09748	0.21176	0.21016	0.98893
KNN	15	0.00000	-0.00203	0.09466	0.09282	0.22222	0.22064	0.98763
KNNW	1	0.00000	-0.00203	0.02395	0.02197	0.08989	0.08804	0.91353
KNNW	18	0.00000	-0.00203	0.07886	0.07699	0.18543	0.18378	0.98788
KNNW	20	0.00000	-0.00203	0.08003	0.07817	0.18072	0.17906	0.98801
KNNW	23	0.00000	-0.00203	0.08105	0.07919	0.17722	0.17554	0.98794
LOF	39	0.15000	0.14827	0.05825	0.05634	0.17391	0.17224	0.96115
LOF	40	0.10000	0.09817	0.05994	0.05803	0.19048	0.18883	0.96116
LOF	41	0.15000	0.14827	0.06171	0.05980	0.19048	0.18883	0.96187
LOF	61	0.10000	0.09817	0.05419	0.05227	0.18605	0.18439	0.96264
SimplifiedLOF	36	0.00000	-0.00203	0.05716	0.05525	0.23077	0.22921	0.95042
SimplifiedLOF	38	0.05000	0.04807	0.05884	0.05693	0.22642	0.22484	0.95187
SimplifiedLOF	57	0.15000	0.14827	0.05499	0.05307	0.17857	0.17690	0.95825
SimplifiedLOF	61	0.15000	0.14827	0.05358	0.05166	0.17857	0.17690	0.95940
LoOP	48	0.00000	-0.00203	0.04942	0.04749	0.19231	0.19067	0.95059
LoOP	59	0.05000	0.04807	0.05170	0.04977	0.18182	0.18016	0.95415
LoOP	60	0.10000	0.09817	0.05155	0.04963	0.18182	0.18016	0.95440
LoOP	82	0.10000	0.09817	0.04889	0.04695	0.18182	0.18016	0.95588
LDOF	29	0.05000	0.04807	0.05604	0.05413	0.09524	0.09340	0.70227
LDOF	77	0.10000	0.09817	0.01441	0.01241	0.10000	0.09817	0.83419
LDOF	80	0.10000	0.09817	0.01450	0.01250	0.10256	0.10074	0.83538
LDOF	85	0.05000	0.04807	0.01405	0.01204	0.09756	0.09573	0.84212
ODIN	6	0.00424	0.00222	0.00325	0.00123	0.00792	0.00590	0.62927
ODIN	100	0.00000	-0.00203	0.03603	0.03407	0.09459	0.09276	0.96531
FastABOD	3	0.10000	0.09817	0.04824	0.04631	0.16667	0.16497	0.82347
FastABOD	98	0.00000	-0.00203	0.02856	0.02658	0.07080	0.06891	0.95920
KDEOS	21	0.05000	0.04807	0.00498	0.00296	0.05128	0.04936	0.58601
KDEOS	33	0.05000	0.04807	0.05303	0.05111	0.09524	0.09340	0.68153
KDEOS	38	0.05000	0.04807	0.05349	0.05157	0.09524	0.09340	0.71560
KDEOS	100	0.05000	0.04807	0.01001	0.00800	0.05128	0.04936	0.85813
LDF	7	0.05000	0.04807	0.05846	0.05655	0.20290	0.20128	0.82476
LDF	12	0.15000	0.14827	0.08384	0.08198	0.19355	0.19191	0.97759
INFLO	23	0.10000	0.09817	0.03648	0.03452	0.16000	0.15829	0.88332
INFLO	56	0.10000	0.09817	0.04481	0.04287	0.15385	0.15213	0.93789
INFLO	64	0.05000	0.04807	0.04311	0.04117	0.17391	0.17224	0.94284
INFLO	100	0.00000	-0.00203	0.03525	0.03330	0.15385	0.15213	0.94699
COF	29	0.15000	0.14827	0.09160	0.08976	0.22222	0.22064	0.96138
COF	38	0.15000	0.14827	0.10417	0.10235	0.19231	0.19067	0.96804
COF	50	0.15000	0.14827	0.14789	0.14616	0.17949	0.17782	0.96549
COF	59	0.20000	0.19838	0.13310	0.13134	0.20000	0.19838	0.96654

Plots

Precision at n

Adjusted precision at n

Average precision

Adjusted average precision

Maximum F1 score

Adjusted maximum F1 score

ROC AUC

Diversity

A: KNN, B: KNNW, C: LOF, D: SimplifiedLOF, E: LoOP, F: LDOF
G: ODIN, H: KDEOS, I: COF, J: FastABOD, K: LDF, L: INFLO