Set-theoretic methodology using fuzzy sets in rule extraction and validation - consistency and coverage revisited

KNOWCON 2020. Knowledge on Economics and Management: Conference Proceedings

“…The investigation of these rules will be performed using the fuzzified set-theoretic consistency and coverage measures (Ragin, 2008) denoted as F1, their modifications proposed by Stoklasa, Luukka and Talášek (2017) denoted F2 (these measures remove the effect ambivalent evidence), F3 (these measures also reflect the effect of pure counterevidence), the degree-of-support and degree-ofdisproof and finally also the F4 measures suggested recently (Stoklasa, Talášek and Luukka, 2018). In all the F1-F3 cases, high consistency can be understood as a support for the given relationship in the data, values of F4 above 0.5 have similar interpretation.…”

“…Ragin, 2008;Schneider and Wagemann, 2014) to identify possible patterns in the data. We will be focusing our analysis mainly on the new fuzzified consistency and coverage measures proposed by Stoklasa, Luukka and Talášek (2017) and Stoklasa, Talášek and Luukka (2018). These measures proved to be useful in social sciences already (Kumbure, Tarkiainen, Luukka et al, 2020) and provide additional insights into the studied relationships in the data by removing ambivalent evidence and counterevidence.…”

Possible drivers of high performance of European mutual ESG funds—an fsQCA view on sustainable investing

“…As the main methodology chosen for this paper is the settheoretic investigation of the consistency of the investigated rules with the data, we will need to introduce the basic (fuzzy) set-theoretic concepts of consistency and coverage as used in the fsQCA [12] and as recently generalized by Stoklasa et. al [5], [6]. We will be employing the revised fuzzification of the consistency and coverage measures [5], [6] as these have already proven useful in practical investigation of reallife relationships in business data [7].…”

“…We therefore apply the tools of the set-theoretic approach and its fuzzification, that are utilized in the frame of the fuzzy set qualitative comparative analysis (fsQCA) -namely we focus on the concepts of the consistency of the rules representing specific assumed relationships with the data and the coverage of these relationships by the available data [3], [4]. Given the recent advances in the methods for fsQCA focusing on the investigation of consistency and coverage of assumed relationships in the fuzzy context, we also apply the recently introduced fuzzified consistency and coverage measures and their alternatives [5], [6]. Another reason to reach for the set-theoretic methods is the fact that based on the definition of the rules (investigated relationships formulated as IF-THEN rules) we can postulate and verify the existence of non-linear relationships between the features of the funds and their performance.…”

Eliciting symptom‐diagnosis knowledge from online medical Q&A

“…How to elicit symptom-diagnosis rules from medical data subject to uncertainty or ambiguity, and how to make a synthesized diagnosis when a Q&A thread is covered by multiple rules? Some promising results on mining text subject to uncertainty have been obtained (Boegl et al, 2004;Chinnnaswamy & Srinivasan, 2018;Malmir et al, 2017;Porebski & Straszecka, 2018;Stoklasa et al, 2017;Sun et al, 2011;Tsipouras et al, 2008;Wang et al, 2016;Wang & Lee, 2011), but with regards to eliciting uncertain information and combining rules, there is still much to explore. Different from the above-mentioned fuzzy expert systems, the knowledge bases in this study are generated within belief rule-based structures so that the diagnostic consequent could reflect slight changes of linguistic variables occurring in the symptom-related text.…”