Stamps (2% of outliers version#03)

A data set representing forged (photocopied or scanned+printed) stamps and genuine (ink) stamps. The features are based on color and printing properties of the stamps. Forged stamps are considered to be outliers. The stamps data set is not taken from the UCI repository, but was used in [1].

References:

[1] B. Micenkova, J. van Beusekom, and F. Shafait. Stamp verification for automated document authentication. In 5th Int. Workshop on Computational Forensics, 2012.

Download all data set variants used (371.2 kB).

Normalized, without duplicates
Not normalized, without duplicates

Normalized, without duplicates

This version contains 9 attributes, 315 objects, 6 outliers (1.90%)

Download raw algorithm results (2.8 MB) Download raw algorithm evaluation table (39.6 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	1	0.00000	-0.01942	0.12273	0.10570	0.25000	0.23544	0.88242
KNN	2	0.00000	-0.01942	0.14357	0.12694	0.30000	0.28641	0.93635
KNN	3	0.00000	-0.01942	0.14403	0.12741	0.31579	0.30250	0.93528
KNNW	1	0.16667	0.15049	0.11337	0.09616	0.25000	0.23544	0.79773
KNNW	2	0.16667	0.15049	0.12093	0.10386	0.26667	0.25243	0.85761
KNNW	5	0.00000	-0.01942	0.13770	0.12095	0.23529	0.22045	0.92503
KNNW	6	0.00000	-0.01942	0.13450	0.11769	0.25000	0.23544	0.92611
LOF	1	0.00000	-0.01942	0.05693	0.03862	0.17391	0.15787	0.59115
LOF	20	0.00000	-0.01942	0.11547	0.09829	0.23529	0.22045	0.91640
LOF	24	0.00000	-0.01942	0.11498	0.09779	0.27273	0.25861	0.91855
SimplifiedLOF	1	0.16667	0.15049	0.08842	0.07072	0.22222	0.20712	0.63916
SimplifiedLOF	37	0.00000	-0.01942	0.10709	0.08975	0.23529	0.22045	0.90992
SimplifiedLOF	47	0.00000	-0.01942	0.10942	0.09212	0.22642	0.21139	0.91100
SimplifiedLOF	54	0.00000	-0.01942	0.10965	0.09237	0.21053	0.19520	0.90992
LoOP	1	0.16667	0.15049	0.08831	0.07061	0.22222	0.20712	0.63592
LoOP	37	0.00000	-0.01942	0.10882	0.09152	0.23077	0.21583	0.91154
LoOP	47	0.00000	-0.01942	0.11162	0.09437	0.21429	0.19903	0.91262
LDOF	3	0.33333	0.32039	0.20586	0.19044	0.44444	0.43366	0.44984
LDOF	47	0.00000	-0.01942	0.10326	0.08585	0.21053	0.19520	0.90399
ODIN	28	0.00000	-0.01942	0.12717	0.11022	0.23529	0.22045	0.92826
ODIN	32	0.16667	0.15049	0.13510	0.11831	0.24000	0.22524	0.92530
ODIN	36	0.05556	0.03722	0.12974	0.11284	0.26087	0.24652	0.92611
FastABOD	3	0.16667	0.15049	0.13365	0.11683	0.30769	0.29425	0.77077
FastABOD	83	0.16667	0.15049	0.11053	0.09326	0.25000	0.23544	0.86192
KDEOS	82	0.16667	0.15049	0.34121	0.32842	0.35294	0.34038	0.95631
KDEOS	84	0.16667	0.15049	0.34169	0.32891	0.31579	0.30250	0.95469
KDEOS	86	0.33333	0.32039	0.33642	0.32353	0.33333	0.32039	0.94391
KDEOS	89	0.33333	0.32039	0.31469	0.30138	0.44444	0.43366	0.94498
LDF	9	0.00000	-0.01942	0.10751	0.09018	0.26667	0.25243	0.91262
LDF	12	0.16667	0.15049	0.12663	0.10967	0.25000	0.23544	0.91532
LDF	33	0.16667	0.15049	0.13082	0.11394	0.25806	0.24366	0.91909
INFLO	1	0.16667	0.15049	0.07504	0.05708	0.20000	0.18447	0.58279
INFLO	23	0.00000	-0.01942	0.11204	0.09480	0.23810	0.22330	0.91532
INFLO	48	0.00000	-0.01942	0.11462	0.09743	0.23810	0.22330	0.91262
INFLO	51	0.00000	-0.01942	0.11127	0.09401	0.24390	0.22922	0.90992
COF	1	0.16667	0.15049	0.08844	0.07074	0.22222	0.20712	0.63916
COF	40	0.16667	0.15049	0.14477	0.12816	0.24390	0.22922	0.92017
COF	41	0.16667	0.15049	0.14329	0.12666	0.23529	0.22045	0.92125
COF	48	0.16667	0.15049	0.13376	0.11694	0.25641	0.24197	0.90022

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

Not normalized, without duplicates