Clustering algorithms for anti-money laundering using graph theory and social network analysis

HEMOLIA (a project under European community’s 7th framework programme) is a new generation Anti-Money Laundering (AML) intelligent multi-agent alert and investigation system which in addition to the traditional financial data makes extensive use of modern society’s huge telecom data source, thereby opening up a new dimension of capabilities to all Money Laundering fighters (FIUs, LEAs) and Financial Institutes (Banks, Insurance Companies, etc.). This Master-Thesis project is done at AIA, one of the partners for the HEMOLIA project in Barcelona. The objective of this thesis is to find the clusters in a network drawn by using the financial data. An extensive literature survey has been carried out and several standard algorithms related to networks have been studied and implemented. The clustering problem is a NP-hard problem and several algorithms like K-Means and Hierarchical clustering are being implemented for studying several problems relating to sociology, evolution, anthropology etc. However, these algorithms have certain drawbacks which make them very difficult to implement. The thesis suggests (a) a possible improvement to the K-Means algorithm, (b) a novel approach to the clustering problem using the Genetic Algorithms and (c) a new algorithm for finding the cluster of a node using the Genetic Algorithm.

No evidence of association between common autoimmunity STAT4 and IL23R risk polymorphisms and non-anterior uveitis.

OBJECTIVE: STAT4 and IL23R loci represent common susceptibility genetic factors in autoimmunity. We decided to investigate for the first time the possible role of different STAT4/IL23R autoimmune disease-associated polymorphisms on the susceptibility to develop non-anterior uveitis and its main clinical phenotypes. METHODS Four functional polymorphisms (rs3821236, rs7574865, rs7574070, and rs897200) located within STAT4 gene as well as three independent polymorphisms (rs7517847, rs11209026, and rs1495965) located within IL23R were genotyped using TaqMan® allelic discrimination in a total of 206 patients with non-anterior uveitis and 1553 healthy controls from Spain. RESULTS No statistically significant differences were found when allele and genotype distributions were compared between non-anterior uveitis patients and controls for any STAT4 (rs3821236: P=0.39, OR=1.12, CI 95%=0.87-1.43; rs7574865: P=0.59 OR=1.07, CI 95%=0.84-1.37; rs7574070: P=0.26, OR=0.89, CI 95%=0.72-1.10; rs897200: P=0.22, OR=0.88, CI 95%=0.71-1.08;) or IL23R polymorphisms (rs7517847: P=0.49, OR=1.08, CI 95%=0.87-1.33; rs11209026: P=0.26, OR=0.78, CI 95%=0.51-1.21; rs1495965: P=0.51, OR=0.93, CI 95%=0.76-1.15). CONCLUSION Our results do not support a relevant role, similar to that described for other autoimmune diseases, of IL23R and STAT4 polymorphisms in the non-anterior uveitis genetic predisposition. Further studies are needed to discard a possible weak effect of the studied variant.

Two functional variants of IRF5 influence the development of macular edema in patients with non-anterior uveitis.

OBJECTIVE Interferon (IFN) signaling plays a crucial role in autoimmunity. Genetic variation in interferon regulatory factor 5 (IRF5), a major regulator of the type I interferon induction, has been associated with risk of developing several autoimmune diseases. In the current study we aimed to evaluate whether three sets of correlated IRF5 genetic variants, independently associated with SLE and with different functional roles, are involved in uveitis susceptibility and its clinical subphenotypes. METHODS Three IRF5 polymorphisms, rs2004640, rs2070197 and rs10954213, representative of each group, were genotyped using TaqMan® allelic discrimination assays in a total of 263 non-anterior uveitis patients and 724 healthy controls of Spanish origin. RESULTS A clear association between two of the three analyzed genetic variants, rs2004640 and rs10954213, and the absence of macular edema was observed in the case/control analysis (P FDR =5.07E-03, OR=1.48, CI 95%=1.14-1.92 and P FDR =3.37E-03, OR=1.54, CI 95%=1.19-2.01, respectively). Consistently, the subphenotype analysis accordingly with the presence/absence of this clinical condition also reached statistical significance (rs2004640: P=0.037, OR=0.69, CI 95%=0.48-0.98; rs10954213: P=0.030, OR=0.67, CI 95%=0.47-0.96), thus suggesting that both IRF5 genetic variants are specifically associated with the lack of macular edema in uveitis patients. CONCLUSION Our results clearly showed for the first time that two functional genetic variants of IRF5 may play a role in the development of macular edema in non-anterior uveitis patients. Identifying genetic markers for macular edema could lead to the possibility of developing novel treatments or preventive therapies.