Gianni Mainetto

A classification-based methodology for planning audit strategies in fraud detection

Planning adequate audit strategies is a key success factor in a posterion’ fraud detection, e.g., in the fiscal and insurance domains, where audits are intended to detect tax evasion and fraudulent claims. A case study is presented in this paper, which illustrates how techniques based on classification can be used to support the task of planning audit strategies. The proposed approach is sensible to some conflicting issues of audit planning, e.g., the trade-off between maximizing audit benefits vs. minimizing audit costs. A methodological scenario, common to a whole class of similar applications, is then abstracted away from the case study. The limitations of available systems to support the identified overall KDD process, bring us to point out the key aspects of a logic-based database language, integrated with mining mechanisms, which is used to provide a uniform, highly expressive environment for the various steps in the construction of the considered case-study.

Static Analysis of Transactions for Conservative Multigranularity Locking

This paper shows a concrete example in which a technique of static analysis, mainly used in the programming language area, can be successfully applied to a database problem. The database problem is the automatic (i.e., without a transaction programmers intervention) realization of a new concurrency control protocol called conservative multiple granularity locking. Being conservative, the scheduler of this protocol ensures that the database resources needed from a transaction are granted before such a transaction begins its execution. Being multigranular, this protocol deals with an hierarchical organization of database resources and it allows to strike a balance between locking overhead and degree of concurrency allowed from one transaction. The analysis we present allows to automatically infer from the text of a transaction a safe approximation of the set of hierarchical database resources needed from the transaction. The analysis gives particular attention to the management of sets of resources to statically foresee if a transaction will access most of the resources in the set. The proposed technique, which can take advantage of statistical information on database resources, infers an approximation close to the actual resources that the transaction is going to use at run time.