The Firm Life Cycle Forecasting Model Using Machine Learning Based on News Articles
- Authors
- Lee, Si Young; Oh, Sae Yong; Lee, Sangwook; Gim, Gwang Yong
- Issue Date
- Jan-2021
- Publisher
- ATLANTIS PRESS
- Keywords
- Firm life cycle prediction; news analysis; machine learning; text feature selection
- Citation
- INTERNATIONAL JOURNAL OF NETWORKED AND DISTRIBUTED COMPUTING, v.9, no.1, pp.1 - 9
- Journal Title
- INTERNATIONAL JOURNAL OF NETWORKED AND DISTRIBUTED COMPUTING
- Volume
- 9
- Number
- 1
- Start Page
- 1
- End Page
- 9
- URI
- http://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/41219
- DOI
- 10.2991/ijndc.k.201218.002
- ISSN
- 2211-7938
- Abstract
- The determination of the firm life cycle has been carried out in relation to the establishment of corporate strategy in the field of accounting or management. The life cycle prediction based on financial information is long because it is determined based on the financial performance of the entity over a year. This study sought to lay the foundation for overcoming this by using news articles to predict the life cycle of a company. In the process of quantifying news article data and predicting the firm life cycle, the method of selecting keywords that can represent the firm life cycle is presented, and the life cycle prediction model is verified with four machine learning techniques using selected candidate keywords. In this study, all four machine learning techniques showed a predicted static classification rate of nearly 60%, demonstrating the availability of news articles, which are unstructured text data, in predicting the corporate life cycle. (C) 2021 The Authors. Published by Atlantis Press B.V.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Business Administration > School of Business Administration > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/ssu/handle/2018.sw.ssu/41219)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.