Application of Neural Network to Business Evaluation
-
- ARISAWA Masaki
- Osaka Institute of Technology
-
- WATADA Junzo
- Osaka Institute of Technology
Bibliographic Information
- Other Title
-
- 経営分析へのニューラルネットワークの適用
- ケイエイ ブンセキ エ ノ ニューラル ネットワーク ノ テキヨウ
Search this article
Abstract
The main purpose of this paper is to build a business evaluation system based on the experts' knowledge and experiences through a neural network which realizes non-linear discrimination and generalization. In the modeling of business evaluation using business evaluation cases, the neural network is taught the input of 5 financial indices and the output of 4 clusters, whose classification is given by experts based on the financial indices. Furthermore, an enhanced method for learning algorithm is proposed in order to shorten the learning time to a larger extent than a conventional method. By its use in the discrimination of samples generated by an exclusive OR and a sine curve, its effectiveness is shown to shorten their learning cycles up to 70% and 50%, respectively. Lastly, we show the effectiveness of the enhanced neural network model for business evaluation to assess the corporations of industrial equipment using real data.
Journal
-
- Journal of Japan Industrial Management Association
-
Journal of Japan Industrial Management Association 47 (5), 307-316, 1996
Japan Industrial Management Association
- Tweet
Details 詳細情報について
-
- CRID
- 1390282680482587392
-
- NII Article ID
- 110003945674
-
- NII Book ID
- AN10561806
-
- ISSN
- 21879079
- 13422618
-
- NDL BIB ID
- 4092539
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- NDL
- CiNii Articles
-
- Abstract License Flag
- Disallowed