Stamps (9% of outliers)

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, 340 objects, 31 outliers (9.12%)

Download raw algorithm results (3.0 MB) Download raw algorithm evaluation table (61.3 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	15	0.25806	0.18363	0.33546	0.26879	0.54206	0.49611	0.90114
KNN	18	0.22581	0.14814	0.33385	0.26702	0.55670	0.51223	0.90051
KNN	74	0.32258	0.25462	0.33597	0.26935	0.50980	0.46063	0.89665
KNN	100	0.32258	0.25462	0.34421	0.27841	0.52000	0.47184	0.89895
KNNW	1	0.25806	0.18363	0.18224	0.10019	0.28571	0.21405	0.67152
KNNW	77	0.22581	0.14814	0.32522	0.25753	0.54386	0.49810	0.89571
KNNW	96	0.25806	0.18363	0.32805	0.26063	0.53097	0.48392	0.89675
KNNW	100	0.25806	0.18363	0.32818	0.26078	0.51786	0.46949	0.89665
LOF	2	0.19355	0.11264	0.12098	0.03279	0.23188	0.15482	0.47970
LOF	100	0.12903	0.04165	0.27272	0.19976	0.40602	0.34642	0.83318
SimplifiedLOF	4	0.19355	0.11264	0.12317	0.03520	0.21053	0.13132	0.50110
SimplifiedLOF	99	0.12903	0.04165	0.21509	0.13634	0.29268	0.22172	0.74350
SimplifiedLOF	100	0.12903	0.04165	0.21516	0.13642	0.29268	0.22172	0.74350
LoOP	5	0.16129	0.07715	0.10943	0.02008	0.17544	0.09272	0.44566
LoOP	93	0.12903	0.04165	0.21737	0.13885	0.29752	0.22705	0.74528
LoOP	100	0.12903	0.04165	0.22058	0.14238	0.29630	0.22570	0.75279
LDOF	4	0.19355	0.11264	0.15448	0.06965	0.25455	0.17976	0.66260
LDOF	100	0.12903	0.04165	0.21966	0.14138	0.29379	0.22294	0.75258
ODIN	30	0.19892	0.11856	0.18257	0.10056	0.29310	0.22219	0.71725
ODIN	40	0.16129	0.07715	0.21647	0.13786	0.29167	0.22060	0.74679
ODIN	100	0.12903	0.04165	0.21629	0.13766	0.31868	0.25033	0.75342
FastABOD	14	0.19355	0.11264	0.17123	0.08809	0.26087	0.18672	0.70947
FastABOD	74	0.12903	0.04165	0.18983	0.10855	0.31429	0.24549	0.75759
FastABOD	97	0.12903	0.04165	0.19069	0.10949	0.31343	0.24455	0.76219
KDEOS	55	0.19355	0.11264	0.13967	0.05336	0.23333	0.15642	0.63618
KDEOS	86	0.09677	0.00616	0.20048	0.12027	0.25503	0.18030	0.68212
KDEOS	91	0.09677	0.00616	0.16594	0.08226	0.26389	0.19004	0.68431
KDEOS	99	0.09677	0.00616	0.16251	0.07850	0.25676	0.18219	0.69130
LDF	96	0.29032	0.21913	0.35356	0.28870	0.51064	0.46154	0.89352
LDF	97	0.29032	0.21913	0.35496	0.29024	0.51613	0.46759	0.89425
LDF	100	0.29032	0.21913	0.34140	0.27533	0.51613	0.46759	0.89550
INFLO	4	0.16129	0.07715	0.11424	0.02538	0.17804	0.09558	0.49107
INFLO	100	0.16129	0.07715	0.24091	0.16476	0.34783	0.28240	0.78923
COF	65	0.25806	0.18363	0.24722	0.17170	0.30303	0.23311	0.75478
COF	93	0.19355	0.11264	0.27916	0.20685	0.39726	0.33679	0.81501
COF	100	0.22581	0.14814	0.24987	0.17461	0.41667	0.35814	0.81867

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