OntoBacen: A Modular Ontology for Risk Management in the Brazilian Financial System.
Filipe Ricardo Polizel, Sara Casare, Jaime Simão Sichman · 2015
This paper presents a first semantic formalization of the Brazilian financial system risk management policies, called OntoBacen, that is based on a modularized approach. We show some partial results generated by a knowledge-based system that uses an ontology constructed to address some domain questions. Episodes like global crisis undermine people’s confidence in the financial system, but also provide lessons for the future. The 2007-2008 meltdown resulted in a significative advancement of governance policies followed by financial institutions worldwide, with some of them treating classic data management problems, as integrity and completeness. To enable the adoption of an integrated and robust global financial system, IT companies and financial institutions are joining efforts for the creation and adoption of a technological framework to better meet the industry needs. The main goal of this work is to explore alternative approaches for the conceptualization and definition of business rules present in governance policies of the Brazilian financial system, more specifically those related to risk management. To this end, it proposes an ontology, called OntoBacen, that expresses the concepts (and their relationships) of this domain, and by using inference algorithms, can verify the compliance of hypothetical financial institutions with those policies. For such a wide and complex domain, modularity must play a central role in the design of the proposed solution, to ensure that it results in an coherent, understandable and scalable knowledge-based system. In the following section, the risk management setting of the Brazilian financial system is briefly introduced. We then present the main initiatives involving ontologies for the financial industry, followed by a description of OntoBacen by means of its properties and requirements. In the sequence, we show the modularization approach adopted by the proposed solution, followed by a section that details how OntoBacen has been developed and how it’s meant to be used. Some test cases illustrate the use of the proposed ontology, followed by current conclusions and future development steps. Financial Industry Risk Management The main good practices for the governance of the global financial system were established and formalized by the Basel Committee on Banking Supervision (BCBS), forming part of the Bank for International Settlements (BIS), that in postcrisis periods identified the need for a robust data management framework. This must ensure that banks have the capacity to aggregate risk exposure data in an integrated manner, reaching all the corporation levels, in addition to standardized risk reporting practices (BIS 2013a), confering the degrees of assertiveness and timeliness required by institutional leaders for decision-making in times of stress. Once established the core principles for banking supervision (BIS 1997; BIS 2012), central banks around the globe have taken them as basis for the establishment of their own regulatory norms. Moreover, they also considered the singularities of their domestic financial systems, what partially mitigated the level of heterogeneity of the global financial system risk management domain. However, this was not sufficient to achieve the goal of representing these domain concepts and their relationships in an integrated manner worldwide, which demands the use of highly integrable and logically grounded tools, such as the use of formal ontologies (Guarino 1995) in the Semantic Web (Berners-Lee, Handler, and Lassila 2001). The policies and guidelines to be followed by the Brazilian financial institutions are created and maintained by the local monetary authorities, but mainly by the executive authority of the national financial system, the Brazil Central Bank, also known as BACEN. In order to align their governance policies with the Basel principles, the Brazilian monetary authorities created a series of norms, known as prudential regulation (BACEN 2014), to be followed by local banks and financial institutions. The main goal of the prudential regulation is to consolidate a national system for risk management and capital adequacy. These regulations are arranged to take into account the main types of risk, described as follows: Credit Risk Associated with the risk of default, the failure to comply with obligations and responsibilities. Market Risk Related to the volatility of rates or prices over the time, such as currency exchange and interest rates, or prices of securities and commodities. Operational Risk Associated with the probability of loss resulting from internal processes failures or deficiencies, including legal risks, such as damages to third parties arising from its activities, or violation of rules established in their jurisdiction. This business division of the domain suggests that modularity should be taken into account (and exploited to the fullest) when dealing with risk management. Ontologies for Financial Industry In recent years, with the greater control over the financial systems by regulatory agencies, the need for information systems interoperability and data integration has increased, which strengthened initiatives related to finance on the Semantic Web; these initiatives, allied with the ontologies’ semantic formalism, have gained their place and importance in this specific industry. As an example, the Suggested Upper Merged Ontology (Niles and Pease 2001), also known as SUMO, has included its own finance domain ontology years ago, dealing with concepts related primarily to financial services, typical of commercial banks, such as bank accounts, payments, loans, etc. A more recent work (in progress at the time of writing), is the Financial Report Ontology1 (FRO), which provides formal and structured meta-information about financial reports, such as balance-sheets; it is primarily based on a well-known XML schema for this application domain, XBRL (Engel et al. 2013), that stands for eXtensible Business Reporting Language. Other relevant initiative is the Financial Industry Business Ontology2 (FIBO), a series of standards being developed by the Enterprise Data Management Council (EDMC) and published following the technical governance process of the Object Management Group (OMG). FIBO currently provides a framework of conceptual definitions concerning the wide spectrum of financial applications, currently available for use in two modules. The first module, FIBO foundations, defines high-level financial concepts such as currency or contracts, and even non-financial concepts such as autonomous agent or country, that are need for the definition of more specific financial concepts. The second module, FIBO Business Entities, defines concepts such as legal persons and corporations, entities that could incur legal obligations such as establishing business contracts with other entities. There are also ontological initiatives concerning financial regulations, where lies the scope of this work. Abi-Lahoud et al. (2013) developed an ontology concerning the compliance with American anti-money-laundering regulations; later, Abi-Lahoud, OBrien, and Butler (2013) presented an experimental discussion about the adopted approach, an iterative process based on subject-matter expertise and on the use of structured natural language, more precisely based on SBVR (OMG 2008), that stands for Semantics of Business Vocabulary and Business Rules, a structured vocabulary founded in formal logic. See: http://xbrl.squarespace.com/financial-report-ontology/. See: http://www.omgwiki.org/OMG-FDTF/doku.php. These initiatives are being conducted with the support of the Governance, Risk and Compliance Technology Centre3 (GRCTC), also responsible for the development of the Financial Industry Regulatory Ontology (FIRO) and the Financial Governance, Risk and Compliance Ontology (FIGO).