scholarly journals A layered rule-based architecture for approximate knowledge fusion?

2010 ◽  
Vol 7 (3) ◽  
pp. 617-642 ◽  
Author(s):  
Barbara Dunin-Kęplicz ◽  
Anh Nguyen ◽  
Andrzej Szałas
Keyword(s):  
Real World ◽  
Description Logics ◽  
Dynamic Logic ◽  
Deductive Database ◽  
General Mechanism ◽  
Multi Agent Systems ◽  
Propositional Dynamic Logic ◽  
Real World Data ◽  
Knowledge Fusion ◽  
Database Technology

In this paper we present a framework for fusing approximate knowledge obtained from various distributed, heterogenous knowledge sources. This issue is substantial in modeling multi-agent systems, where a group of loosely coupled heterogeneous agents cooperate in achieving a common goal. In paper [5] we have focused on defining general mechanism for knowledge fusion. Next, the techniques ensuring tractability of fusing knowledge expressed as a Horn subset of propositional dynamic logic were developed in [13,16]. Propositional logics may seem too weak to be useful in real-world applications. On the other hand, propositional languages may be viewed as sublanguages of first-order logics which serve as a natural tool to define concepts in the spirit of description logics [2]. These notions may be further used to define various ontologies, like e.g. those applicable in the Semantic Web. Taking this step, we propose a framework, in which our Horn subset of dynamic logic is combined with deductive database technology. This synthesis is formally implemented in the framework of HSPDL architecture. The resulting knowledge fusion rules are naturally applicable to real-world data.

scholarly journals Tractable approximate knowledge fusion using the Horn fragment of serial propositional dynamic logic

2010 ◽  
Vol 51 (3) ◽  
pp. 346-362 ◽  
Author(s):  
Barbara Dunin-Ke¸plicz ◽  
Linh Anh Nguyen ◽  
Andrzej Szałas
Keyword(s):  
Dynamic Logic ◽  
Propositional Dynamic Logic ◽  
Knowledge Fusion ◽  
Horn Fragment

The Influence of the Test Operator on the Expressive Power of PDL-like Logics

2019 ◽  
Vol 29 (8) ◽  
pp. 1289-1310
Author(s):  
Linh Anh Nguyen
Keyword(s):  
Description Logics ◽  
Dynamic Logic ◽  
Expressive Power ◽  
Propositional Dynamic Logic ◽  
Test Operator ◽  
Local Reflexivity ◽  
Horn Fragment ◽  
Logical Formalism

Abstract Berman and Paterson proved that test-free propositional dynamic logic (PDL) is weaker than PDL. One would raise questions: does a similar result also hold for extensions of PDL? For example, is test-free converse-PDL (CPDL) weaker than CPDL? In what circumstances the test operator can be eliminated without reducing the expressive power of a PDL-based logical formalism? These problems have not yet been studied. As the description logics $\mathcal{ALC}_{trans}$ and $\mathcal{ALC}_{reg}$ are, respectively, variants of test-free PDL and PDL, there is a concept of $\mathcal{ALC}_{reg}$ that is not equivalent to any concept of $\mathcal{ALC}_{trans}$. Generalizing this, we prove that there is a concept of $\mathcal{ALC}_{reg}$ that is not equivalent to any concept of the logic that extends $\mathcal{ALC}_{trans}$ with inverse roles, nominals, qualified number restrictions, the universal role and local reflexivity of roles. We also provide some results for the case with RBoxes and TBoxes. One of them states that tests can be eliminated from TBoxes of the deterministic Horn fragment of $\mathcal{ALC}_{reg}$.

Decidability and Complexity of Some Finitely-valued Dynamic Logics

2021 ◽  
Author(s):  
Igor Sedlár
Keyword(s):  
Transitive Closure ◽  
Description Logics ◽  
Dynamic Logic ◽  
Time Algorithm ◽  
Action Execution ◽  
Propositional Dynamic Logic ◽  
Relational Models ◽  
Reasoning About Actions ◽  
Decidability And Complexity ◽  
Dynamic Logics

Propositional Dynamic Logic, PDL, is a well known modal logic formalizing reasoning about complex actions. We study many-valued generalizations of PDL based on relational models where satisfaction of formulas in states and accessibility between states via action execution are both seen as graded notions, evaluated in a finite Łukasiewicz chain. For each n>1, the logic PDŁn is obtained using the n-element Łukasiewicz chain, PDL being equivalent to PDŁ2. These finitely-valued dynamic logics can be applied in formalizing reasoning about actions specified by graded predicates, reasoning about costs of actions, and as a framework for certain graded description logics with transitive closure of roles. Generalizing techniques used in the case of PDL we obtain completeness and decidability results for all PDŁn. A generalization of Pratt's exponential-time algorithm for checking validity of formulas is given and EXPTIME-hardness of each PDŁn validity problem is established by embedding PDL into PDŁn.

Abstract #946: Assessment of Acromegaly Patients with and Without Diabetes Treated with Lanreotide Depot: 2-Year Real World Data

2016 ◽  
Vol 22 ◽  
pp. 219
Author(s):  
Roberto Salvatori ◽  
Olga Gambetti ◽  
Whitney Woodmansee ◽  
David Cox ◽  
Beloo Mirakhur ◽  
...  
Keyword(s):  
Real World ◽  
Real World Data ◽  
World Data

Safety and efficacy of direct acting oral anticoagulants and vitamin K antagonists in nonvalvular atrial fibrillation – a network meta-analysis of real-world data

VASA ◽  
2019 ◽  
Vol 48 (2) ◽  
pp. 134-147 ◽  
Author(s):  
Mirko Hirschl ◽  
Michael Kundi
Keyword(s):  
Real World ◽  
Meta Analysis ◽  
Vitamin K Antagonists ◽  
Oral Anticoagulants ◽  
Efficacy And Safety ◽  
Nonvalvular Atrial Fibrillation ◽  
Real World Data ◽  
Hazard Ratios ◽  
Direct Acting ◽  
Event Rates

Abstract. Background: In randomized controlled trials (RCTs) direct acting oral anticoagulants (DOACs) showed a superior risk-benefit profile in comparison to vitamin K antagonists (VKAs) for patients with nonvalvular atrial fibrillation. Patients enrolled in such studies do not necessarily reflect the whole target population treated in real-world practice. Materials and methods: By a systematic literature search, 88 studies including 3,351,628 patients providing over 2.9 million patient-years of follow-up were identified. Hazard ratios and event-rates for the main efficacy and safety outcomes were extracted and the results for DOACs and VKAs combined by network meta-analysis. In addition, meta-regression was performed to identify factors responsible for heterogeneity across studies. Results: For stroke and systemic embolism as well as for major bleeding and intracranial bleeding real-world studies gave virtually the same result as RCTs with higher efficacy and lower major bleeding risk (for dabigatran and apixaban) and lower risk of intracranial bleeding (all DOACs) compared to VKAs. Results for gastrointestinal bleeding were consistently better for DOACs and hazard ratios of myocardial infarction were significantly lower in real-world for dabigatran and apixaban compared to RCTs. By a ranking analysis we found that apixaban is the safest anticoagulant drug, while rivaroxaban closely followed by dabigatran are the most efficacious. Risk of bias and heterogeneity was assessed and had little impact on the overall results. Analysis of effect modification could guide the clinical decision as no single DOAC was superior/inferior to the others under all conditions. Conclusions: DOACs were at least as efficacious as VKAs. In terms of safety endpoints, DOACs performed better under real-world conditions than in RCTs. The current real-world data showed that differences in efficacy and safety, despite generally low event rates, exist between DOACs. Knowledge about these differences in performance can contribute to a more personalized medicine.

Sign in / Sign up

Export Citation Format

Share Document