ARTIFICIAL INTELLIGENCE THE PROCESSING OF ARCHIVAL DOCUMENTS

Enfoque

INTRODUCTION

The main impact of these social transformations is the increase in the production of information, this increase in volume combined with the possibilities of applying statistical analysis and data science, brings with it new questions that reveal a set of opportunities for applications and a vast interdisciplinarity between Information Science and Data Science.

One of the recent challenges for information scientists and data scientists has been to extract relevant information from large amounts of unstructured data, such as texts written in multiple languages. The main approach to text and language analysis through computational means is called Natural Language Processing, a technique that can be a great ally of information professionals to provide the classification of documents that were still produced in paper format but are indexed in databases.

Our scope covers an investigation on the possibility of using Artificial Intelligence, specifically Natural Language Processing, can be used by Information Science, Archives and Libraries as a tool for the treatment of data, information and knowledge that are the object of this science. This article also describes in a general way the procedures that are being carried out by the Municipality of São Caetano do Sul aiming at the use of a database of indexed documents to carry out the documentary classification according to the archival criteria.

Although there is a set of studies in the international scientific literature focusing on Information Science and Artificial Intelligence, case studies dealing with the use of technology for the treatment of archival documents are incipient. In the Brazilian context, through a search in the Brapci database, no article was identified that dealt with the use of artificial intelligence for the treatment of archival collections.

METHODOLOGY

The study presented in this article was carried out through a bibliographic, exploratory, and qualitative research. Allowing to carry out in the literature the review of concepts that are fundamental to deal with artificial intelligence to treat archival documents.

CONCLUSIONS

It is possible to observe that there is a growth in the adoption of solutions that use Artificial Intelligence as a basis in various sectors of the economy and this will be no different in the fields of study of Information Science (Archives, Libraries and Museums). This area, recognizing the disciplinary diversity that the data have acquired, focuses mainly on its scientific transdisciplinary and on the fact that it constitutes a scientific novelty, following both the needs of the labor market and the epistemological needs of the academic environment. Thus, it can be a great ally in the processes of management, preservation, and use of information.

It is more common to identify studies that focus on the use of AI in the context of Information Science as a tool only for the treatment of information created in a digital environment. However, although the case study presented is in its initial phase, it is possible to identify that this technology can also be used for the treatment of analog collections. The possibility of automating the work of document classification reveals the need for information professionals to acquire new skills for the job market, considering a new technological context.