A Multi-Attribute Approach to Knowledge Representation for Loan Granting

Abstract

The complexity of finenciel decision-making problems is such that automation of the reasoning process by conventionel approaches is often incomplete or inadequete. This paper describes CREDEX. s knowledge-based system which is being developed to assist bank-loan officers in interpreting end evaluating the activities of firms applying for a loan. CREDEX is written in SNARK. It integrates shallow and deep knowledge through e multi-level structure driven by e mete-expert. The system builds on psychological research on informetion processing and handles risk assessment through a combination of four multi-attribute models. I

Cite

Text

Pinson. "A Multi-Attribute Approach to Knowledge Representation for Loan Granting." International Joint Conference on Artificial Intelligence, 1987.

Markdown

[Pinson. "A Multi-Attribute Approach to Knowledge Representation for Loan Granting." International Joint Conference on Artificial Intelligence, 1987.](https://mlanthology.org/ijcai/1987/pinson1987ijcai-multi/)

BibTeX

@inproceedings{pinson1987ijcai-multi,
  title     = {{A Multi-Attribute Approach to Knowledge Representation for Loan Granting}},
  author    = {Pinson, Suzanne},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {1987},
  pages     = {588-591},
  url       = {https://mlanthology.org/ijcai/1987/pinson1987ijcai-multi/}
}