Comparing Alternative Storage Models for Words Extracted from Legal Texts

Abstract

The COVID-19 pandemic created new demands for services in the judicial system, requiring the use of a data warehouse (DW). Although there exist approaches that use DW in the judicial domain, few target the pandemic or publicly provide the information extracted from the texts. Following the needs of a legal expert, we have developed the COVID-19 Portal. It extracts documents from the Supreme Federal Court in Brazil to obtain quantitative information on words used in the texts. In this paper, we present the design of a DW, and show the query performance improvement achieved with its implementation. The DW has been developed on Postgres, and its performance is compared with the original implementation on MongoDB Cloud and a local MongoDB database.