基于时态数据库的审计方法体系构建技术研究

发布时间：2018-08-12 17:31

【摘要】：作为审计业务的数据基础,采集的数据必须能够完整正确地反应被审计对象的历史状态。目前的审计业务数据主要存在两个问题:第一,没有将历史数据与当前数据联系起来,使得作为审计方法执行基础的审计业务数据不完整,造成审计结果不完整;第二,数据被更新后,新的数据就会覆盖历史数据,更新操作一旦执行,历史数据就被删除,无法找回。且目前的审计方法并不能够有效的利用时态属性进行审计。基于时态数据库的审计方法体系构建分为两个部分:审计数据和审计方法。本文利用时态数据库理论,在关系数据库中对审计数据内容和审计数据表结构进行时态重构。利用实体完整性约束条件,将审计数据表中的主属性与时间属性共同作为新表的主属性,通过“多表合一”技术,解决了相同关系中1个实体只有1条记录以及数据冗余的问题,从而将历史数据与当前数据相结合;利用时态数据库理论,对审计数据表结构进行有效时间维和事务时间维的扩展,对有效时间维的扩展,解决了因数据更新,历史数据被覆盖的问题,对事务时间维的扩展使审计人员能够了解数据生产时间,对审计结果追踪提供了帮助。并通过在“多表合一”前后的数据集上执行相同的审计方法的对比,说明了时态重构审计数据内容能够有效地将历史数据与当前数据相结合,使审计结果更完整。在审计数据时态重构的基础上,本文提出了时间数据类型处理方法和WHERE子句处理方法,并将两个方法结合,提出审计方法时态重构算法,该算法能够在不改变审计方法原有语义的情况下,更加全面有效地利用时态属性进行审计工作,使得审计结果更加精确。将该算法应用于现有的审计方法中,验证了经过时态重构后的审计方法能够正确执行,并且没有改变原审计方法的审计目的,满足了审计人员关注时间属性的审计需求。
[Abstract]:As the data base of audit business, the collected data must be able to reflect the historical state of the audited object completely and correctly. There are two main problems in the current audit business data: first, not linking the historical data with the current data, which makes the audit business data, which is the basis of the audit method implementation, incomplete, resulting in incomplete audit results; second, When the data is updated, the new data overwrites the historical data. Once the update operation is performed, the historical data is deleted and cannot be retrieved. And the current audit method can not effectively use temporal attributes to audit. The construction of audit method system based on temporal database is divided into two parts: audit data and audit method. In this paper, the content of audit data and the structure of audit data table are reconstructed in relational database by using temporal database theory. The main attribute and the time attribute in the audit data table are taken as the main attributes of the new table by using the constraint condition of entity integrity. It solves the problem that one entity has only one record and data redundancy in the same relationship, which combines the historical data with the current data, and makes use of the theory of temporal database. The structure of audit data table is extended to the dimension of effective time and the dimension of transaction time, and the extension of the dimension of effective time solves the problem that the historical data is overwritten because of the updating of the data. The extension of transaction time dimension enables auditors to understand data production time and helps audit result tracking. By comparing the same audit methods on the data sets before and after "multi-table integration", it is shown that the temporal reconstruction audit data can effectively combine the historical data with the current data, and make the audit results more complete. On the basis of temporal reconfiguration of audit data, this paper proposes a method of processing time data type and a method of processing WHERE clause, and combines the two methods to propose an algorithm of temporal refactoring of audit method. Without changing the original semantics of the audit method, the algorithm can make more comprehensive and effective use of temporal attributes to audit, and make the audit results more accurate. The algorithm is applied to the existing audit methods, and it is verified that the audit method after temporal reconstruction can be executed correctly, and the audit purpose of the original audit method has not been changed, which meets the audit needs of the auditors concerned about the time attribute.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.13

【参考文献】