Fuzzy knowledge discovery and decision-making through clustering and Dynamic tables : Application in Colombian business Finance

Abstract

This article aims to analyze financially the companies belonging to the Colombian services sector through the Altman Z Score model and to make a new model based on the use of fuzzy logic. The methodology for the development of fuzzy models based on clustering and pivot tables was used for the development of the fuzzy inference model. The data were obtained from the Legis Comex database on financial information for the year 2017 for the services sector of Colombia. In total, there were 5,372 companies. The type of problem for this case study is classification. Results show 90% of accuracy, with an average absolute deviation of 0.096 (less than 10%) with a Kappa index of 84.7%, indicating excellent performance in the task of classifying a company's bankruptcy risk. As a main conclusion we can establish that the implemented methodology could be considered a learning technique because it uses multiple layers of processing that help represent the behavior of data with multiple levels of abstraction, helping to convert linear problems into nonlinear due to the uniqueness of fuzzy logic.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.