Case-based reasoning on loan negotiation | |
| Author | Tamisin, Mairene R. |
| Call Number | AIT Thesis no. CS-91-22 |
| Subject(s) | Reasoning Problem solving--Data processing |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Abstract | The principle of Case-based Reasoning is based on what we people do from day to day. People normally depend on past experiences encountered to deal with a new problem being faced. Loan negotiation is a specific domain of interest applicability of case-based reasoning is investigated. negotiations are carried on not solely based on clear-cut but mainly on subjective human decisions, there is a need Expert System that simulates loan experts in the way they and make decisions during the negotiation process. where Since rules, for an behave Experiences - good and bad - have to provide Problem Solvers learning apart from knowledge. A good mechanism therefore, for storing all past experiences and remembering the most appropriate and similar case to a new problem encountered is applied, through the conceptual graph representation of cases and other background knowledge. And a menu-driven prototype system embodying this mechanism is designed and implemented in Prolog, for a Company-Bank loan negotiation based on the methodology of case-based reasoning. Furthermore, it can be proved that as the number of cases of problem-solving increases, the system' s learning also increases, which may either be caused from inside the System or from the outside world which constantly changes and acqui res new knowledge. Support modules are also developed to facilitate the learning process of the Case-based Reasoner. One is the mechanism that takes care of duplicate and near-duplicate cases in the case memory, and this is done through a method of least generalization. Other modules are designed to update a successful case to a failure case when it fails, to check loan status of an applicant, and to compute and evaluate negotiation payoff values on the bank's point of view. All of these supporting modules together with the Case-based Reasoner made up the Case-based Reasoning System which can serve as an advisor of loan experts. |
| Year | 1991 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Sadananda, Ramakot |
| Examination Committee(s) | Vilas Wuwongse ;Soegaard, Ragnar |
| Scholarship Donor(s) | The Government of Australia |
| Degree | Thesis (M.Sc.) - Asian Institute of Technology, 1991 |