2018
DOI: 10.1016/j.asoc.2017.10.037
|View full text |Cite
|
Sign up to set email alerts
|

Predicting corporate investment/non-investment grade by using interval-valued fuzzy rule-based systems—A cross-region analysis

Abstract: Systems for predicting corporate rating have attracted considerable interest in soft computing research due to the requirements for both accuracy and interpretability. In addition, the high uncertainty associated primarily with linguistic uncertainties and disagreement among experts is another challenging problem. To overcome these problems, this study proposes a hybrid evolutionary interval-valued fuzzy rule-based system, namely IVTURS, combined with evolutionary feature selection component. This model is use… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
10
0
1

Year Published

2018
2018
2024
2024

Publication Types

Select...
4
3
2

Relationship

3
6

Authors

Journals

citations
Cited by 25 publications
(11 citation statements)
references
References 74 publications
0
10
0
1
Order By: Relevance
“…However, unlike the other typical type of multi-model systems, namely, ANNs [17], most of the EISs are designed for processing streaming data "on the fly", and they are able to self-update and self-evolve their system structure and meta-parameters to follow the rapidly changing data patterns of data streams [29]. Currently, EISs have been successfully implemented for various real-world applications including classification [33], prediction [20], control [34], anomaly detection [27], etc. The most popular (neuro-) fuzzy systems include, but not limited to, eTS [2], DENFIS [22], eClass [6], SAFIS [36], PANFIS [31], GENIFS [32] and IT2FNN [41].…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…However, unlike the other typical type of multi-model systems, namely, ANNs [17], most of the EISs are designed for processing streaming data "on the fly", and they are able to self-update and self-evolve their system structure and meta-parameters to follow the rapidly changing data patterns of data streams [29]. Currently, EISs have been successfully implemented for various real-world applications including classification [33], prediction [20], control [34], anomaly detection [27], etc. The most popular (neuro-) fuzzy systems include, but not limited to, eTS [2], DENFIS [22], eClass [6], SAFIS [36], PANFIS [31], GENIFS [32] and IT2FNN [41].…”
Section: Related Workmentioning
confidence: 99%
“…For each newly arrived data sample, , it is firstly sent to the massively parallel fuzzy rules, and the local decision-maker will identify the most similar prototype to within this fuzzy rule and calculate the score of confidence using equation (20):…”
Section: Validation Processmentioning
confidence: 99%
“…In this study, we evaluated both the number of antecedents (conditions in the rules) and the size of rule base (number of rules). It should also be noted that trade-off between accuracy and interpretability have been observed in earlier research and it is the user that should provide his/her preferences (Hajek, 2018). We decided to prefer the interpretability of the models at the granularity level and, therefore the number of linguistic terms was fixed to five for all models.…”
Section: Resultsmentioning
confidence: 99%
“…These can also be adapted in future studies to raise accuracy. Finally, other methods such as association rules [11], [25] and other evolutionary rule learning or selection [26], [27] can be used to generate membership and non-membership functions and better fit the variability of the patterns in the data.…”
Section: Discussionmentioning
confidence: 99%