Optimizing the Data Loss Prevention Level Using Logic Paraconsistent Annotated Evidential Eτ

Resumo (EN)

Currently, corporations worldwide have a problem that grows exponentially: orchestrating the organization and understanding structured and unstructured data. These unstructured data can be grouped in repositories, for isolated and random data entry, however on the other hand the data loss analysis, that is, data loss prevention, where some criteria of artificial intelligence create templates that are monitored, and that, because they are very restricted, also present contradictions and flaws. The focus of this study is to optimize this analysis by minimizing the level of data loss using the Logic Paraconsistent Annotated Evidential Eτ. With a bibliographic review on DLP—Data Loss Prevention, Logic Paraconsistent Annotated Evidential Eτ, Artificial Intelligence techniques, and data protection. With the use of a Python program, and applied research will be carried out with data from a financial company, which presents 40% of data loss in its analysis with this process of Artificial Intelligence, in comparison with the minimization of this data loss with the use of Logic Paraconsistent Annotated Evidential Eτ to 25%, that is, a difference of 15%.