Comparing Text Mining Algorithms for Predicting Irregularities in Public Accounts

Authors

Breno Santana Santos, UFS - Federal University of SergipeFollow
Methanias Colaço Júnior, UFS – Federal University of SergipeFollow
Bruno Cruz da Paixão, UFS - Federal University of SergipeFollow
Rafael M. Santos, UFS - Federal University of SergipeFollow
André Vinicius Rodrigues P Nascimento, UFS – Federal University of SergipeFollow
Hallan Cosmo dos Santos, Court of Accounts of Sergipe - TCE-SEFollow
Wallace H. L. Filho, Court of Accounts of Sergipe - TCE-SEFollow
Arquimedes S. L. de Medeiros, Court of Accounts of Sergipe - TCE-SEFollow

Document Type

Article

Publication Date

5-2015

Keywords

Text Mining, Public Accounts History, Court of Accounts, Audit

Abstract

Information systems that support public sector daily activities generate large data sets. As a large proportion of the data in these data sets are text, Text Mining can play an important role in deriving potentially useful and previously unknown information. The overall goal of this paper is evaluate the performance and quality of three text mining classification algorithms applied to detect irregularities in public sector records. To evaluate the algorithms, a tool was designed and a case study was carried out at the Court of Accounts of Sergipe. Performance and Quality metrics were evaluated: mean execution time, accuracy, precision, coverage and F-measure. The results show that the multinomial naive bayes algorithm using inverse document frequency was the best approach to find evidences of travel reimbursement irregularities.

Comments

This paper is in Portuguese (Análise Comparativa de Algoritmos de Mineração de Texto Aplicados a Históricos de Contas Públicas)

Recommended Citation

Santos, Breno Santana; Colaço, Methanias Júnior; da Paixão, Bruno Cruz; Santos, Rafael M.; Nascimento, André Vinicius Rodrigues P; dos Santos, Hallan Cosmo; Filho, Wallace H. L.; and de Medeiros, Arquimedes S. L., "Comparing Text Mining Algorithms for Predicting Irregularities in Public Accounts" (2015). Proceedings of the XI Brazilian Symposium on Information Systems (SBSI 2015). 12.
https://aisel.aisnet.org/sbis2015/12