Inference and reasoning in a Bayesian knowledge-intensive CBR system

Hoda Nikpour,Agnar Aamodt

doi:10.1007/s13748-020-00223-1

Abstract

This paper presents the inference and reasoning methods in a Bayesian supported knowledge-intensive case-based reasoning (CBR) system called BNCreek. The inference and reasoning process in this system is a combination of three methods. The semantic network inference methods and the CBR method are employed to handle the difficulties of inferencing and reasoning in uncertain domains. The Bayesian network inference methods are employed to make the process more accurate. An experiment from oil well drilling as a complex and uncertain application domain is conducted. The system is evaluated against expert estimations and compared with seven other corresponding systems. The normalized discounted cumulative gain (NDCG) as a rank-based metric, the weighted error (WE), and root-square error (RSE) as the statistical metrics are employed to evaluate different aspects of the system capabilities. The results show the efficiency of the developed inference and reasoning methods.

Highlights

The main role of the inference and reasoning process in AI systems is interpreting raw data to generate new information from the domain knowledge
8.2.2 Results In Fig. 6, we report on normalized discounted cumulative gain (NDCG) at four ranks { cut@5, cut@10, cut@15, cut@20 }
In order to make inferencing on the data, acquiring knowledge, representing the obtained knowledge, and reasoning on it, semantic network and Bayesian network analysis are employed

Summary

Introduction

The main role of the inference and reasoning process in AI systems is interpreting raw data to generate new information from the domain knowledge. It is not possible to create a complete model from an uncertain domain. Different inference and reasoning methods are utilized for working in uncertain domains, which are weak-theory domains in the sense that relationships between concepts are uncertain. Statements derived from within domain models are uncertain. The contrast would be perfect-theory domains, in which relations are certain, and statements can be proved true or false. This lack of certainty means that in order to build a representative knowledge model, more knowledge is needed to support

Results

Discussion

Conclusion