Open AccessClassification of the Priority of Auditing XBRL Instance Documents with Fuzzy Support Vector Machines Algorithm
Author(s) -
Guang-Yih Sheu
Publication year - 2019
Publication title -
journal of autonomous intelligence
Language(s) - English
Resource type - Journals
ISSN - 2630-5046
DOI - 10.32629/jai.v2i2.55
Subject(s) - xbrl , support vector machine , computer science , conformity , fuzzy logic , data mining , algorithm , audit , variable (mathematics) , machine learning , accounting , artificial intelligence , mathematics , business , world wide web , law , political science , mathematical analysis
Concluding the conformity of XBRL (eXtensible Business Reporting Language) instance documents law to the Benford's law yields apparently different results before and after a company's financial distress. These results bring an idea of finding fraudulent documents from the inspection of financial ratios since the unacceptable conformity implies a large likelihood of a fraudulent document. Fuzzy support vector machines models are developed to implement such an idea. The dependent variable is a fuzzy variable quantifying the conformity of an XBRL instance document to the Benford's law; whereas, independent variables are financial ratios. Nevertheless, insufficient data are available to define any membership function for describing the fuzziness in independent and dependent variables, but the interval factor method is introduced to express that fuzziness. Using the resulting fuzzy support vector machines model, it is suggested that the price-to-book ratio versus equity ratio may be used to classify the priority of auditing XBRL instance documents. The misclassification rate is less than 30 \%. In conclusion, a new and promising application of fuzzy support vector machines algorithm has been found in this study.