**6. Conclusions and Future Work**

We presented an extension of the DAQAP platform to incorporate probabilistic knowledge bases, giving rise to the P-DAQAP system, which is, to the best of our knowledge, the first system of its kind for probabilistic defeasible reasoning. After discussing the details of its design and describing applications to cybersecurity, we performed an empirical evaluation whose goal it was to explore the effectiveness and efficiency of world sampling-based approximate query answering. Our study showed that the entropy associated with the probability distribution over worlds has a large impact on expected solution quality, but even a modest number of samples suffices to reach good-quality approximations. Compared to classical (nonprobabilistic) approaches, the results of our experiments show that P-DAQAP allows for representing, effectively and efficiently reasoning with different types of uncertainty, modeling complex domains in more detail, and providing more informed answers that can be accompanied by explanations. In critical environments, having outputs of this kind increases credibility and trust in the system by its users.

Future work involves carrying out a broader evaluation investigating other sampling methods, avoiding repeated samples, and testing other probabilistic models. One of the goals of this research line is to develop a method to guide knowledge engineering efforts on the basis of domain features, requirements in terms of expressive power, approximation quality, and query response time.

**Author Contributions:** Conceptualization, M.A.L., A.J.G., P.S. and G.I.S.; methodology, M.A.L., P.S. and G.I.S.; validation, M.A.L., P.S. and G.I.S.; formal analysis, G.I.S.; investigation, M.A.L. and G.I.S.; writing—original draft, M.A.L.; writing—review and editing, M.A.L., A.J.G., P.S. and G.I.S.; project administration, G.I.S.; supervision and funding acquisition, A.J.G. and G.I.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Universidad Nacional del Sur (UNS) grant numbers PGI 24/ZN34 and PGI 24/N046, Universidad Nacional de Entre Ríos grant number PDTS-UNER 7066, and Agencia Nacional de Promoción Científica y Tecnológica, Argentina grant number grants PICT-2018-0475 (PRH-2014-0007). P.S. is supported by internal funding from the ASU Fulton Schools of Engineering.

**Institutional Review Board Statement:** Not Applicable.

**Informed Consent Statement:** Not Applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

The following abbreviations are used in this manuscript:

