Evaluation of fuzzy membership functions for linguistic rule-based classifier focused on explainability, interpretability and reliability

Porębski, Sebastian

doi:10.1016/j.eswa.2022.117116

Cited by 19 publications

(4 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To address the uncertainty in classification problems, many valuable methods have been proposed. Porebski 8 propose a new technique of linguistic rule extraction, which adopts a fuzzy membership function to describe the imprecision of linguistic values and measured the uncertainty by a fuzzy confidence function. Yang et al 9 establish a rule-based system named Cumulative Belief Rule-Based System to overcome the limitation of the classical rule-based system.…”

Section: Introductionmentioning

confidence: 99%

A new correlation belief function in Dempster-Shafer evidence theory and its application in classification

Tang

Zhang

Zhou

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Uncertain information processing is a key problem in classification. Dempster-Shafer evidence theory (D-S evidence theory) is widely used in uncertain information modelling and fusion. For uncertain information fusion, the Dempster’s combination rule in D-S evidence theory has limitation in some cases that it may cause counterintuitive fusion results. In this paper, a new correlation belief function is proposed to address this problem. The proposed method transfers the belief from a certain proposition to other related propositions to avoid the loss of information while doing information fusion, which can effectively solve the problem of conflict management in D-S evidence theory. The experimental results of classification on the UCI dataset show that the proposed method not only assigns a higher belief to the correct propositions than other methods, but also expresses the conflict among the data apparently. The robustness and superiority of the proposed method in classification are verified through experiments on different datasets with varying proportion of training set.

show abstract

Section: Introductionmentioning

confidence: 99%

A new correlation belief function in Dempster-Shafer evidence theory and its application in classification

Tang

Zhang

Zhou

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Bilgi çizgesi ile ilgili diğer bir gereksinim ise bilgiler arası sıralamanın sağlanabilmesidir [30]. Bu gereksinimin sağlanabilmesi için çalışmamızda, bilgi çizgesine güvenilirlik mekanizması entegre edilmiştir.…”

Section: Introductionunclassified

Novel Extensions to the Knowledge Graph Using the Hashing Mechanism

TAKAN

2023

International Journal of Advances in Engineering and Pure Sciences

View full text Add to dashboard Cite

Bilginin doğasına ilişkin, onu şekillendiren çok çeşitli unsurlar bulunmaktadır. Örneğin güvenirlik, tutarlılık, değişmezlik ve bağlam gibi mekanizmalar bunların başında gelir. Ancak söz konusu mekanizmaların bilgi çizgesinde temsil edilmesi oldukça yaygın bir problemdir. Çalışmamızda bu problemin çözümüne katkıda bulunmak amacıyla, bilginin karmaşık doğasına ilişkin güven, tutarlılık, değişmezlik ve bağlam gibi temel mekanizmalar, hashing teknolojisi kullanılarak bilgi çizgesine entegre edilmiştir. Çalışmamızda bu eklentiler, bilgi çizgesinden ayrı tutularak, yapıların işlevselliklerinin bozulmaması sağlanmıştır. Geliştirdiğimiz eklentiler sayesinde bir bilgi değiştiğinde onu etkileyen tüm bilgilerin otomatik güncellenmesi, belirsizlik, bilgiler arasında sıralama yapılamaması, bazı bilgilerin değişmez olarak tutulamaması ve bilgiler arasında hızlı bir karşılaştırmanın yapılamaması gibi yaygın bilgi çizgesi problemleri, örnek senaryolar üzerinden test edilerek çözüme kavuşturulmuştur. Çalışmamızın, bilgi çizgesinin iyileştirilmesine yönelik literatüre ve bilgi çizgesini kullanan yapay zeka yazılımlarının geliştirilmesine katkı sunması beklenmektedir.

show abstract

“…Another requirement for the KG is to ensure the ordering of knowledge ( Porebski, 2022 ). We have integrated a reliability mechanism into the KG to fulfill this requirement.…”

Section: Introductionmentioning

confidence: 99%

Knowledge graph augmentation: consistency, immutability, reliability, and context

Takan¹

2023

PeerJ Computer Science

View full text Add to dashboard Cite

A knowledge graph is convenient for storing knowledge in artificial intelligence applications. On the other hand, it has some shortcomings that need to be improved. These shortcomings can be summarised as the inability to automatically update all the knowledge affecting a piece of knowledge when it changes, ambiguity, inability to sort the knowledge, inability to keep some knowledge immutable, and inability to make a quick comparison between knowledge. In our work, reliability, consistency, immutability, and context mechanisms are integrated into the knowledge graph to solve these deficiencies and improve the knowledge graph’s performance. Hash technology is used in the design of these mechanisms. In addition, the mechanisms we have developed are kept separate from the knowledge graph to ensure that the functionality of the knowledge graph is not impaired. The mechanisms we developed within the scope of the study were tested by comparing them with the traditional knowledge graph. It was shown graphically and with t-test methods that our proposed structures have higher performance in terms of update and comparison. It is expected that the mechanisms we have developed will contribute to improving the performance of artificial intelligence software using knowledge graphs.

show abstract

Evaluation of fuzzy membership functions for linguistic rule-based classifier focused on explainability, interpretability and reliability

Cited by 19 publications

References 69 publications

A new correlation belief function in Dempster-Shafer evidence theory and its application in classification

A new correlation belief function in Dempster-Shafer evidence theory and its application in classification

Novel Extensions to the Knowledge Graph Using the Hashing Mechanism

Knowledge graph augmentation: consistency, immutability, reliability, and context

Contact Info

Product

Resources

About