Extended Belief Rule-Base Optimization Base on Clustering Tree and Parameter Optimization

Jinhui Zhuang,Jifeng Ye,Nannan Chen,Yanggeng Fu,Weijie Fang,Xuecheng Fan

doi:10.1109/access.2021.3051001

Jinhui Zhuang, Jifeng Ye + Show 4 more

Open Access

https://doi.org/10.1109/access.2021.3051001

Copy DOI

Journal: IEEE Access	Publication Date: Jan 1, 2021
Citations: 26	License type: CC BY 4.0

Affiliation: Fuzhou University, Sichuan University

Abstract

Extended belief rule-based (EBRB) system has a better ability to model complex problems than belief rule-based (BRB) system. However, the storage of rules in EBRB system is out of order, which leads to the low efficiency of rule retrieval during the reasoning process. Therefore, to improve the efficiency of rule retrieval, this study introduces K-means clustering tree algorithm into the construction of rule base, then proposes a multi-layer weighted reasoning approach based on K-means clustering tree. The proposed approach seeks out a path on the tree during the rule retrieval process, and then figures out several reasoning results according to the nodes on the path. These results are weighted and aggregated to obtain the final conclusion of the system, thus ensure both the efficiency of reasoning and the sufficient utilization of information. In addition, the differential evolution (DE) algorithm is used to train the parameters of EBRB system in this study. Several experiments are conducted on commonly used classification datasets from UCI, and the results are compared with some existing works of EBRB system and conventional machine learning methods. The comparison results illustrate that the proposed method can make an obvious improvement in the performance of EBRB system.

Highlights

In order to effectively handle the uncertain quantitative and qualitative information and model a complex decision-making problem, Professor Yang et al put forward a belief rule-based (BRB) system [1], which is based on D-S evidence theory [2], [3], fuzzy theory [4], decision theory [5] and IF- rule base [6]
It can be seen that the reasoning accuracy of MKTDE-extended belief rule-based (EBRB) is effectively improved as the iteration times of the differential evolution (DE) algorithm increasing, which with an increase of 4%
The comparison results are shown in Table 4: It can be seen from Table 4 that MKTDE-EBRB proposed in this paper has better reasoning accuracy than the others methods in Ecoli, Iris, and Glass, which gains an increase of 2.35%, 0.67% and 2.10% than the second-best results, respectively

Summary

INTRODUCTION

In order to effectively handle the uncertain quantitative and qualitative information and model a complex decision-making problem, Professor Yang et al put forward a belief rule-based (BRB) system [1], which is based on D-S evidence theory [2], [3], fuzzy theory [4], decision theory [5] and IF- rule base [6]. The BRB system has aroused widespread concern [7]–[9] and has been well applied in many fields, such as regional railway safety assessment [10], bridge risk assessment [11] and sensor network health assessment [12] On this basis, Liu et al [13] embedded the belief distribution to the antecedent term of rules and proposed the extended belief rule-based (EBRB) system. The major contributions include: 1) It is innovatively proposed to combine the K-means clustering tree with the EBRB system to construct a new structure of rule base to achieve the efficient rule retrieval in reasoning. Otherwise the rule is incomplete, and the value of βjk needs to be modified using (2): Tk

CONSTRUCTION OF EBRB

REASONING OF EBRB

CONSTRUCTION OF K-MEANS CLUSTERING TREE

PARAMETER TRAINING BASED ON DIFFERENTIAL EVOLUTION ALGORITHM

EXPERIMENT AND ANALYSIS

Findings

CONCLUSION