Treffer: پ ی شب ی ن ی درماند گ ی ما ل ی شرک تها با استفاده از اطلاعات مت ن ی گزار شه ا ی فعا ل ی ت ه یئ تم د یر ه

Objective: Financial distress is a serious concern for the economic sustainability of countries. The individual and social costs associated with financial distress have made its prediction a crucial issue for many managers, stakeholders, policymakers, and auditors. While most prior studies have relied on structured and quantitative financial data, this research aims to employ text mining techniques and machine learning algorithms to predict financial distress using unstructured data extracted from board of directors' activity reports. Method: To achieve this objective, the board of directors' reports of 100 listed companies over the period 2011-2021 were collected. The textual content of these reports was analyzed using Python, involving preprocessing, feature extraction, and feature selection. the statistical analysis and modeling phase was carried out using various machine learning algorithms. Results: The results of the study indicated the superiority of the two methods, the decision tree model and the support vector machine with radial kernel, over other methods (including logistic regression, random forest, nearest neighbor, and support vector machine methods with linear, sigmoid, and polynomial kernels). Conclusion: The findings of this study suggest that, rather than solely relying on numerical data and the ratios derived from such figures, text mining techniques can also be effectively utilized for analysis and prediction. Furthermore, by integrating insights from unstructured textual data with structured quantitative information, it is possible to enhance the prediction of corporate financial distress. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Accounting Knowledge is the property of Shahid Bahonar University of Kerman, Faculty of Management & Economics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)