On the Evaluation of Classification Methods Applied to Requests for Revision of Registered Debts

Lima, Helton Souza; Fernandes, Damires Yluska de Souza; Moura, Thiago; Sabóia, Daniel

doi:10.5220/0010498403350342

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These works confirm improvements in decision support applied to tax administration processes. Lima et al [16] applied cost-insensitive methods to the R3D classification problem. This work achieved promising results with the cost-insensitive Random Forest model w.r.t.…”

Section: Related Workmentioning

confidence: 99%

“…Regarding the works which make use of tax administration datasets, none of them use cost-sensitive classification methods [24,14,17,5]. When taking the R3D classification problem into account [16], the achieved learning model would take the risk of granting a request involving millions of Brazilian reals in debts that should not be forgiven. At last, although Mehta et al, [20] face a tax administration problem using cost-sensitive classification methods, they focus on tax evasion.…”

Section: Related Workmentioning

confidence: 99%

“…Not only to support human decisions on the processes but also to reduce the costs of debt analyses, thus enabling faster and more assertive decisions, some supervised classification models have been evaluated to indicate the likelihood for an R3D to be approved or rejected [16]. [16] evaluated some cost-insensitive learning algorithms regarding important and traditional measures w.r.t. the problem of provinding predictions on R3Ds.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the evaluation of example-dependent cost-sensitive models for tax debts classification

Lima¹,

Fernandes²,

Moura³

2022

Anais Do XIX Encontro Nacional De Inteligência Artificial E Computacional (ENIAC 2022)

View full text Add to dashboard Cite

Example-dependent cost-sensitive classification methods are suitable to many real-world classification problems, where the costs, due to misclassification, vary among every example of a dataset. Tax administration applications are included in this segment of problems, since they deal with different values involved in the tax payments. To help matters, this work presents an experimental evaluation which aims to verify whether cost-sensitive learning algorithms are more cost-effective on average than traditional ones. This task is accomplished in a tax administration application domain, what implies the need of a cost-matrix regarding debt values. The obtained results show that cost-sensitive methods avoid situations like erroneously granting a request with a debt involving millions of reals. Considering the savings score, the cost-sensitive classification methods achieved higher results than their traditional method versions.

show abstract

Section: Related Workmentioning

confidence: 99%