Stamps (2% of outliers version#07)

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 (2.7 MB) Download raw algorithm evaluation table (36.4 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	0.16667	0.15049	0.20493	0.18949	0.34483	0.33211	0.94822
KNN	2	0.16667	0.15049	0.23105	0.21612	0.38710	0.37520	0.95793
KNNW	1	0.16667	0.15049	0.17164	0.15556	0.28571	0.27184	0.87109
KNNW	6	0.16667	0.15049	0.22122	0.20609	0.37500	0.36286	0.95631
LOF	1	0.16667	0.15049	0.07027	0.05222	0.21053	0.19520	0.59682
LOF	6	0.16667	0.15049	0.17636	0.16037	0.30769	0.29425	0.91532
LOF	11	0.16667	0.15049	0.29670	0.28305	0.28571	0.27184	0.93905
LOF	12	0.16667	0.15049	0.29473	0.28104	0.28571	0.27184	0.93959
SimplifiedLOF	1	0.16667	0.15049	0.06021	0.04196	0.18182	0.16593	0.49407
SimplifiedLOF	9	0.16667	0.15049	0.20287	0.18739	0.30769	0.29425	0.90885
SimplifiedLOF	10	0.16667	0.15049	0.28933	0.27553	0.30769	0.29425	0.91640
SimplifiedLOF	21	0.16667	0.15049	0.28509	0.27121	0.28571	0.27184	0.93042
LoOP	1	0.16667	0.15049	0.05992	0.04167	0.18182	0.16593	0.48544
LoOP	6	0.16667	0.15049	0.22699	0.21199	0.28571	0.27184	0.75782
LoOP	23	0.16667	0.15049	0.27857	0.26456	0.28571	0.27184	0.92988
LoOP	39	0.16667	0.15049	0.27863	0.26462	0.28571	0.27184	0.92665
LDOF	5	0.16667	0.15049	0.07019	0.05213	0.20000	0.18447	0.61435
LDOF	13	0.16667	0.15049	0.22198	0.20687	0.28571	0.27184	0.82740
LDOF	47	0.16667	0.15049	0.28656	0.27270	0.28571	0.27184	0.93258
ODIN	9	0.16667	0.15049	0.07682	0.05890	0.16667	0.15049	0.82848
ODIN	36	0.16667	0.15049	0.27634	0.26229	0.28571	0.27184	0.92961
ODIN	48	0.16667	0.15049	0.28009	0.26611	0.28571	0.27184	0.92853
FastABOD	3	0.16667	0.15049	0.13791	0.12117	0.25000	0.23544	0.87918
FastABOD	12	0.16667	0.15049	0.21870	0.20353	0.28571	0.27184	0.93689
FastABOD	15	0.16667	0.15049	0.22674	0.21172	0.28571	0.27184	0.94121
FastABOD	19	0.16667	0.15049	0.22202	0.20691	0.26667	0.25243	0.94175
KDEOS	71	0.33333	0.32039	0.22813	0.21315	0.36364	0.35128	0.94714
KDEOS	77	0.33333	0.32039	0.23212	0.21721	0.36364	0.35128	0.95092
KDEOS	95	0.33333	0.32039	0.29961	0.28601	0.44444	0.43366	0.93959
LDF	3	0.33333	0.32039	0.24715	0.23253	0.40000	0.38835	0.91963
LDF	8	0.16667	0.15049	0.17357	0.15752	0.30303	0.28950	0.94229
LDF	11	0.16667	0.15049	0.29624	0.28257	0.28571	0.27184	0.93905
INFLO	1	0.16667	0.15049	0.10142	0.08398	0.25000	0.23544	0.46926
INFLO	7	0.16667	0.15049	0.22707	0.21206	0.28571	0.27184	0.82578
INFLO	22	0.16667	0.15049	0.27619	0.26214	0.28571	0.27184	0.92071
COF	11	0.16667	0.15049	0.22648	0.21146	0.42105	0.40981	0.94606
COF	24	0.33333	0.32039	0.32541	0.31231	0.33333	0.32039	0.94013

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