Showing papers on "Abductive reasoning published in 2018"

Explainable AI: The new 42?

Abstract: Explainable AI is not a new field. Since at least the early exploitation of C.S. Pierce’s abductive reasoning in expert systems of the 1980s, there were reasoning architectures to support an explanation function for complex AI systems, including applications in medical diagnosis, complex multi-component design, and reasoning about the real world. So explainability is at least as old as early AI, and a natural consequence of the design of AI systems. While early expert systems consisted of handcrafted knowledge bases that enabled reasoning over narrowly well-defined domains (e.g., INTERNIST, MYCIN), such systems had no learning capabilities and had only primitive uncertainty handling. But the evolution of formal reasoning architectures to incorporate principled probabilistic reasoning helped address the capture and use of uncertain knowledge.

Perspective—Discovery Within Validation Logic: Deliberately Surfacing, Complementing, and Substituting Abductive Reasoning in Hypothetico-Deductive Inquiry

Abstract: We propose a more explicit role for abductive reasoning, or the development of initial explanation, in hypothetico-deductive (H-D) inquiry. We begin by describing the roots of abduction in pragmati...

Truth-Seeking by Abduction

Abstract: In a seminar with the title “Deduction and Induction in the Sciences”, it is intriguing to ask the following questions: (i) Is there a third type of inference besides deduction and induction? (ii) Does this third type of inference play a significant role within scientific inquiry? A positive answer to both of these questions was advocated by Charles S. Peirce throughout his career, even though his opinions changed in important ways during the fifty years between 1865 and 1914. Peirce called the third kind of inference “hypothesis”, “abduction”, or “retroduction”.1 In this paper, I shall follow Peirce’s steps in discussing abduction by analyzing its logical form (Sections 1–3), its role in science (Section 4), and the grounds of its validity (Sections 5–6). We shall see that Peirce’s discussion is more insightful than many recent attempts to analyze abductive inference. Still, recently some progress has been made in the treatment of abduction within the Bayesian theory of epistemic probability (Section 6) and truth-approximation (Section 7). The results of this work support the view of scientific realism: abduction or inference to the best explanation (IBE), combined with empirical and experimental testing of scientific theories, is the best method of seeking informative truths in science (Section 8).

Bayesian model reduction

Abstract: This paper reviews recent developments in statistical structure learning; namely, Bayesian model reduction. Bayesian model reduction is a method for rapidly computing the evidence and parameters of probabilistic models that differ only in their priors. In the setting of variational Bayes this has an analytical solution, which finesses the problem of scoring large model spaces in model comparison or structure learning. In this technical note, we review Bayesian model reduction and provide the relevant equations for several discrete and continuous probability distributions. We provide worked examples in the context of multivariate linear regression, Gaussian mixture models and dynamical systems (dynamic causal modelling). These examples are accompanied by the Matlab scripts necessary to reproduce the results. Finally, we briefly review recent applications in the fields of neuroimaging and neuroscience. Specifically, we consider structure learning and hierarchical or empirical Bayes that can be regarded as a metaphor for neurobiological processes like abductive reasoning.

A Review of Mixed Methods, Pragmatism and Abduction Techniques

Abstract: The purpose of this paper is to propose that mixed methods research is complementary to traditional qualitative or quantitative research, also that pragmatism is an attractive philosophical partner for mixed methods. A key feature of mixed methods research is its methodological pluralism that can lead to superior research. The research question is whether ‘pragmatism’ as a philosophical choice to combine positivism and interpretivism can lead to an appreciation of 'what works' in practice? (Tashakkori & Teddlie, 2010). The paper posits that pragmatism supports the use of different research methods and that a continuous cycle of inductive, deductive and when appropriate, abductive reasoning, produces useful knowledge and serves as a rationale for rigorous research. Firstly, the so called “paradigm wars” of quantitative or qualitative analysis are briefly reviewed; and the tenets of pragmatism are explained. A comparison is made of the different approaches and the value of applying abduction techniques to ‘surprising facts or puzzles’. Secondly, the literature regarding the ubiquity of abduction techniques is explored. Third, two recent empirical case studies in the airline and engineering sectors are summarised. Abductive thinking was key to explaining empirical phenomenon relating to competition, and in particular how leading UK and German multinationals developed rather different approaches to outsourcing. Finally, in conclusion, mixed methods were found to combine numerical and cognitive reasoning that led to a ‘best answer’ to data that otherwise could not be adequately explained. Furthermore, the application of different approaches can lead to research and subsequent management decisions that reflect both the interplay of social and scientific aspects of the world today.

Abductive reasoning as the basis to reproduce expert criteria in ECG Atrial Fibrillation identification

Abstract: Objective: This work aims at providing a new method for the automatic detection of atrial fibrillation, other arrhythmia and noise on short single lead ECG signals, emphasizing the importance of the interpretability of the classification results Approach: A morphological and rhythm description of the cardiac behavior is obtained by a knowledge-based interpretation of the signal using the \textit{Construe} abductive framework Then, a set of meaningful features are extracted for each individual heartbeat and as a summary of the full record The feature distributions were used to elucidate the expert criteria underlying the labeling of the 2017 Physionet/CinC Challenge dataset, enabling a manual partial relabeling to improve the consistency of the classification rules Finally, state-of-the-art machine learning methods are combined to provide an answer on the basis of the feature values Main results: The proposal tied for the first place in the official stage of the Challenge, with a combined $F_1$ score of 083, and was even improved in the follow-up stage to 085 with a significant simplification of the model Significance: This approach demonstrates the potential of \textit{Construe} to provide robust and valuable descriptions of temporal data even with significant amounts of noise and artifacts Also, we discuss the importance of a consistent classification criteria in manually labeled training datasets, and the fundamental advantages of knowledge-based approaches to formalize and validate that criteria

Abductive reasoning as a basis to reproduce expert criteria in ECG atrial fibrillation identification.

Abstract: Objective This work aims at providing a new method for the automatic detection of atrial fibrillation, other arrhythmia and noise on short single-lead ECG signals, emphasizing the importance of the interpretability of the classification results. Approach A morphological and rhythm description of the cardiac behavior is obtained by a knowledge-based interpretation of the signal using the Construe abductive framework. Then, a set of meaningful features are extracted for each individual heartbeat and as a summary of the full record. The feature distributions can be used to elucidate the expert criteria underlying the labeling of the 2017 PhysioNet/CinC Challenge dataset, enabling a manual partial relabeling to improve the consistency of the training set. Finally, a tree gradient boosting model and a recurrent neural network are combined using the stacking technique to provide an answer on the basis of the feature values. Main results The proposal was independently validated against the hidden dataset of the Challenge, achieving a combined F 1 score of 0.83 and tying for the first place in the official stage of the Challenge. This result was even improved in the follow-up stage to 0.85 with a significant simplification of the model, attaining the highest score so far reported on the hidden dataset. Significance The obtained results demonstrate the potential of Construe to provide robust and valuable descriptions of temporal data, even with the presence of significant amounts of noise. Furthermore, the importance of consistent classification criteria in manually labeled training datasets is emphasized, and the fundamental advantages of knowledge-based approaches to formalize and validate those criteria are discussed.

Understanding Clinical Reasoning from Multiple Perspectives: A Conceptual and Theoretical Overview

Abstract: Rather than a historical overview as in Chap. 2, this chapter provides the reader with insight into the various approaches that have been used to understand clinical reasoning. We review concepts and major scholars who have been involved in such investigations. Cognitive psychologists Newel and Simon theorized about problem-solving skills and artificial intelligence and initiated the use of computers as metaphors of thinking. Elstein and colleagues found that there is no such thing as a general problem-solving skill, independent of medical knowledge, and thus clinical reasoning is case specific. Reasoning then became analyzed in approaches, including forward reasoning from data to diagnosis; hypothetico-deductive reasoning with backward nature, from hypothesis to diagnosis; and abductive reasoning to understand early hypothesis generation that is so characteristic in clinical reasoning, elaborated by Patel and colleagues. Bordage introduced prototypes to characterize how physicians may remember illness presentations and semantic qualifiers to denote the shortened conceptual language and labels physicians use to store medical information systematically in memory. Illness scripts represent how encounters with diseases are remembered by physicians and were introduced by Feltovich and Barrows. Schmidt and Boshuizen elaborated the concept further and propose encapsulation of knowledge as a hypothetical process that happens when physicians regularly and routinely apply shortcuts in thinking typically ellaborated as pathophysiology. Reasoning ability appears not only to be case specific-- it is also situation or context specific. Clinicians with broad reasoning ability have extensive experience. Deliberate practice with many cases and in varying contexts is recommended by Ericsson to acquire reasoning expertise. To improve reasoning, some authors have focused on cognitive biases and error prevention. Norman, however, concludes that bias reduction strategies are unlikely to be successful but correcting knowledge deficiencies is likely to lead to reasoning success. Kahnemann promoted System 1 and System 2 thinking for instant pattern recognition (nonanalytic reasoning) and analytic reasoning, respectively. What actually happens in the brain during clinical reasoning is the domain of neuroscience, which may provide insights from research in the near future.

A Peircean account of concepts: grounding abstraction in phylogeny through a comparative neuroscientific perspective.

Abstract: The nature of concepts has always been a hotly debated topic in both philosophy and psychology and, more recently, also in cognitive neuroscience. Different accounts have been proposed of what concepts are. These accounts reflect deeply different conceptions of how the human mind works. In the last decades, two diametrically opposed theories of human cognition have been discussed and empirically investigated: the Computational Theory of Mind, on the one hand (Fodor 1983 The modularity of mind: an essay on faculty psychology ; Pylyshyn 1984 Computation and cognition: toward a foundation for cognitive science ), and Embodied Cognition (Barsalou 2008 Annu. Rev. Psychol. 59 , 617–645. (doi:10.1146/annurev.psych.59.103006.093639); Gallese & Lakoff 2005 Cogn. Neuropsychol. 22 , 455–479 (doi:10.1080/02643290442000310); Shapiro 2011 Embodied cognition ), on the other hand. The former proposes that concepts are abstract and amodal symbols in the language of thought, while the latter argues for the embodied nature of concepts that are conceived of as grounded in actions and perception. The embodiment of both concrete and abstract concepts has been challenged by many (e.g. Mahon & Caramazza 2008 J. Physiol. 102 , 59–70. (doi:10.1016/j.jphysparis.2008.03.004); Caramazza et al . 2014 Annu. Rev. Neurosci. 37 , 1–15. (doi:10.1146/annurev-neuro-071013-013950)). These challenges will be here taken seriously and addressed from a comparative perspective. We will provide a phylogenetic and neurobiologically inspired account of the embodied nature of both abstract and concrete concepts. We will propose that, although differing in certain respect, they both might have a bodily foundation. Commonalities between abstract and concrete concepts will be explained by recurring to the Peircean notions of icon and abductive inference (CP 2.247). According to Peirce, icons are the kind of signs on which abductive inferences rest (Peirce CS 1931 in Collected papers of Charles S. Peirce, Hartshorne C, Weiss P, Burks AW. (eds), 40; Peirce CS 1997 In The 1903 Harvard lectures on pragmatism (ed. A. Turrisi)). It will be claimed that the mechanism of Embodied Simulation (Gallese & Sinigaglia 2011 Trends Cogn. Sci. 15 , 512–519. (doi:10.1016/j.tics.2011.09.003)) can be described as an icon (Cuccio V & Caruana F. 2015 Il corpo come icona. Abduzione, strumenti ed Embodied Simulation. Versus, n. 119, 93–103), and it will then be suggested that on these, basic natural signs rest, both phylogenetically and ontogenetically, the capacity to conceptualize. This article is part of the theme issue ‘Varieties of abstract concepts: development, use and representation in the brain’.

Leaping into real-world relevance: an ‘abduction’ process for nonprofit research

Abstract: Positioned in the midst of the heated debate about the production of relevant and usable knowledge for practitioners in the nonprofit sector and a serious shortage of high-impact research that speaks to practice, the purpose of this Research Note is to direct nonprofit scholarship toward embracing “abduction,” which is the initial creative stage in scientific inquiry that facilitates the formulation of testable explanatory hypotheses and makes new discoveries in a sensory and logically structured way. We use an emerging interest in social innovation by the nonprofit sector as an illustrative example to show the advantages of using abductive reasoning as the primary method of reasoning for discovering new knowledge of a nascent but vital phenomenon. The novel contribution of this Research Note lies in encouraging scholarship on the nonprofit sector to an applied “practice-led” research process that is intellectually relevant and has the potential to bridge the scholar–practice divide.

Critical historical research method and marketing scholarship

Abstract: As a research toolkit grounded in the philosophy and practice of critical analysis, critical ‘historical research method’ (critical HRM) can help explain marketing-related phenomena that shape and are shaped by social context and human agency The value to marketing scholarship of critical HRM, which relies on critical realism (CR) and abductive reasoning (AR), is shown by (1) presenting elements of effective historical research, and (2) arguing historical research that entails social factors and their interactions compels a methodology for capturing the social and dynamic features of research objects Implications of critical HRM for theory building and marketing practice are then summarised

What Are They Thinking?: Eliciting Student Reasoning About Troublesome Concepts in Introductory Computer Science

Abstract: Understanding student conceptions and identifying student misconceptions is an important precursor to developing high quality pedagogical materials and approaches. We conducted a pencil-and-paper survey consisting of a demographic questionnaire and 21 content-based questions with 106 students at the start of a second semester course for CS majors (i.e., CS2) at a large public U.S. university with a focus on engineering and science. We then selected the questions that appeared to be most problematic for students and several closely related questions and analyzed the student reasoning for each question and across questions with similar concepts. We found that students exhibited misconceptions related to pass by value versus pass by reference semantics, "false sharing" of variables of the same name in different scopes, and believing that global variables may not be accessed from the main function and/or that a global variable may be modified via a local variable of the same name (i.e., in the presence of variable shadowing). Potential explanations for these misconceptions include fragile knowledge, lack of attention to detail, over-attention to newly acquired knowledge, information foraging (abductive reasoning) in the presence of misleading contextual clues, and lack of exposure to the use of global variables beyond the specification of constants.

A novel approach for analysing evolutional motivation of empirical engineering knowledge

Abstract: Empirical engineering knowledge (EEK), a specific technical know-how about solving engineering problems, is frequently accumulated and reused in this era of mass innovation and knowledge-driven eco...

Teaching as Abductive Reasoning: The Role of Argumentation

Abstract: The view that argumentation is a desired reasoning practice in the classroom is well reported in the literature Nonetheless, it is still not clear what type of reasoning supports classroom argumentation The paper discusses abductive reasoning as the most adequate for students’ arguments to emerge in a classroom discussion Abductive reasoning embraces the idea of plausibility and defeasibility of both the premises and the conclusion As such, teachers’ role becomes the one of guiding students through formulating relevant hypotheses and selecting the most plausible one according to criteria Argumentation schemes are proposed as useful tools in this processL'idee que l'argumentation est une pratique de raisonnement souhaitee en classe est bien documentee dans la litterature Neanmoins, il n'est toujours pas clair quel type de raisonnement soutient l'argumentation en classe Dans cet article on discute du raisonnement abductif comme etant le plus adequat pour que les arguments des eleves emergent dans une discussion en classe Le raisonnement abductif emploie l'idee de plausibilite et de la revocabilite des premisses et de la conclusion En tant que tel, le role des enseignants consiste a guider les eleves a formuler des hypotheses pertinentes et a selectionner le plus plausible selon des criteres Les schemes d'argumentation sont proposes comme des outils utiles dans ce processus

The Role of Abduction in Production of New Ideas in Design

Abstract: The pragmatist philosopher Peirce insisted that besides deduction and induction there is a third main form of inference, abduction, which is the only type of inference capable of producing new ideas. Also he defined abduction as a stage of the methodological process in science, where hypotheses are formed to explain anomalies. Basing on these seminal ideas, scholars have proposed modified, widened or alternative definitions of abduction and devised taxonomies of abductive inferences. Influenced by Peirce’s seminal writings and subsequent treatments on abduction in philosophy of science, design scholars have in the last 40 years endeavoured to shed light on design by means of the concept of abduction. The first treatment was provided by March in 1976. He viewed that abduction, which he called “productive reasoning”, is the key mode of reasoning in design. He also presented a three-step cyclic design process, similar to Peirce’s methodological process in science. Among the many other later treatments of design abduction, Roozenburg’s definition of explanatory and innovative abduction is noteworthy. However, an evaluation of the related literature suggests that research into abduction in design is still in an undeveloped stage. This research shows gaps in coverage, lack of depth and diverging outcomes. By focusing on the differences between science and design as well as on empirical knowledge of different phenomena comprising design, new conceptions of abduction in design are derived. Given the differences of context, abduction in design shows characteristics not yet found or identified in science. For example, abduction can occur in connection to practically all inference types in design; it is a property of an inference besides an inference itself. A number of the most important abductive inference types as they occur in design are identified and discussed in more detail.

Peirce on Abduction

Abstract: As a philosophical term in English, ‘abduction’ was originally a seventeenth century translation of a Latin term used by Renaissance logicians. It was adopted in 1898 by Charles S. Peirce who gave it a significant role in his system. Peirce (1839–1914) was the founder of American pragmatism, well-known and appreciated for his studies in formal logic and probabilistic reasoning. He argued already in 1865 that, besides deduction and induction, there is a third type of inference which he called ’hypothesis’, ‘presumption’, ‘retroduction’, or ‘abduction’. This is reasoning from effects to causes or from surprising observations to explanatory theories. Peirce’s account of this ampliative inference changed in important ways during the 50 years between 1865 and 1914. After Peirce’s death, his ideas gradually become known with the publication of six volumes of his Collected Papers (CP) in 1931–1935. Peirce’s conception of abductive reasoning became a hot topic in the philosophy of science after World War II, when N. R. Hanson suggested that abduction is a logic of discovery, Gilbert Harman argued that all types of inductive reasoning can be reduced to inference to the best explanation (IBE), Howard Smokler analyzed abduction as an important method of confirmation, and Larry Laudan treated abduction as an instruction for the pursuit of a hypothetical theory. In this chapter, we shall follow Peirce’s steps in discussing abduction by analyzing its logical and probabilistic forms (Sects. 1.1, 1.2, and 1.3), its main interpretations (Sect. 1.4), and some current debates about Peirce’s distinctions (Sect. 1.5).

Computer Tools and Techniques for Lawyers and the Judiciary

Abstract: This overview of computing for law considers the perspective of legal firms, before turning to differences of jurisdiction, to regulatory changes, to implications for staffing and access to justice. After a taxonomy of applications, a section on procedural support systems is followed with ones on discovery, on predicting the likely outcome of litigation, on argumentation, on case-based automated reasoning and abductive reasoning (dealing in a subsection with descriptive vs prescriptive software for modeling or supporting sentencing), and then to various artificial intelligence approaches to legal evidence. Before concluding, we consider the Bayesian controversy among legal evidence scholars.

Eco and Peirce on Abduction

Abstract: This paper argues that Umberto Eco had a sophisticated theory of abductive reasoning and that this theory is fundamentally akin to Peirce’s both in the analysis and in the justification of this kind of reasoning The first section expounds the essentials of Peirce’s theory of abduction, and explains how Peirce moved from seeing abduction as a kind of reasoning to seeing it as a stage of the larger process of inquiry The second section deals with one of Eco’s paradigmatic examples of abduction, ie, William of Baskerville’s abduction concerning the horse Brunellus in the overture of The Name of the Rose, and shows that, just like in Peirce’s three-stages model of inquiry, William’s abductions are verified by means of deduction and induction The third section examines the problem of the justification of abductive reasoning, and argues that both Peirce and Eco solved this problem through the idea that the justification of abduction is itself abductive (meta-abduction in Eco, ur-abduction in Peirce)

Weight learning in a probabilistic extension of answer set programs

Abstract: LPMLN is a probabilistic extension of answer set programs with the weight scheme derived from that of Markov Logic. Previous work has shown how inference in LPMLN can be achieved. In this paper, we present the concept of weight learning in LPMLN and learning algorithms for LPMLN derived from those for Markov Logic. We also present a prototype implementation that uses answer set solvers for learning as well as some example domains that illustrate distinct features of LPMLN learning. Learning in LPMLN is in accordance with the stable model semantics, thereby it learns parameters for probabilistic extensions of knowledge-rich domains where answer set programming has shown to be useful but limited to the deterministic case, such as reachability analysis and reasoning about actions in dynamic domains. We also apply the method to learn the parameters for probabilistic abductive reasoning about actions.

Abductive Reasoning in Mathematics Education: Approaches to and Theorisations of a Complex Idea

Abstract: There is a growing literature on the importance of abductive reasoning in mathematics education. However, there are some important variations in what exactly is referred to as ‘abductive reasoning’. This article identifies key theorisations of abductive reasoning in the work of Peirce, Habermas and Eco, as well as describing different approaches to abductive reasoning found the mathematics education literature. A framework in which the different approaches taken in the research literature can be placed and compared is proposed.

Thinking in Patterns and the Pattern of Human Thought as Contrasted with AI Data Processing

Abstract: We propose that the ability of humans to identify and create patterns led to the unique aspects of human cognition and culture as a complex emergent dynamic system consisting of the following human traits: patterning, social organization beyond that of the nuclear family that emerged with the control of fire, rudimentary set theory or categorization and spoken language that co-emerged, the ability to deal with information overload, conceptualization, imagination, abductive reasoning, invention, art, religion, mathematics and science. These traits are interrelated as they all involve the ability to flexibly manipulate information from our environments via pattern restructuring. We argue that the human mind is the emergent product of a shift from external percept-based processing to a concept and language-based form of cognition based on patterning. In this article, we describe the evolution of human cognition and culture, describing the unique patterns of human thought and how we, humans, think in terms of patterns.

The Power of Ethnography: A Useful Approach to Researching Politics

Abstract: This article explores and argues the suitability of ethnographic methods, primarily participant observation, in the research of politics, and in the construction of concepts for theory building. Th...

Legal Reasoning and Argumentation

Abstract: Wigmore thought that there was a science of proof underlying legal reasoning that could be displayed in any given case as a graphic sequence of argumentation from the evidence in the case leading to the ultimate probandum. Argumentation technology has now vindicated this approach by providing useful qualitative methods that can be applied to identifying, analyzing, and evaluating the pro and con arguments put forward by both sides in a trial. In this chapter, it is shown how to apply argumentation schemes, such as argument from witness testimony, using argument diagrams applied to typical sequences of evidence-based legal reasoning in which there is a successive refinement of arguments displayed as the evidence comes into a case. It is shown how argument from applying rules to cases, teleological (goal-directed reasoning), argument from expert witness testimony, and abductive reasoning (represented as inference to the best explanation) are centrally important for this purpose.

Challenges Facing Evidence-Based Prevention: Incorporating an Abductive Theory of Method.

Abstract: Current systems used to determine whether prevention programs are "evidence-based" rely on the logic of deductive reasoning. This reliance has fostered implementation of strategies with explicitly stated evaluation criteria used to gauge program validity and suitability for dissemination. Frequently, investigators resort to the randomized controlled trial (RCT) combined with null hypothesis significance testing (NHST) as a means to rule out competing hypotheses and determine whether an intervention works. The RCT design has achieved success across numerous disciplines but is not without limitations. We outline several issues that question allegiance to the RCT, NHST, and the hypothetico-deductive method of scientific inquiry. We also discuss three challenges to the status of program evaluation including reproducibility, generalizability, and credibility of findings. As an alternative, we posit that extending current program evaluation criteria with principles drawn from an abductive theory of method (ATOM) can strengthen our ability to address these challenges and advance studies of drug prevention. Abductive reasoning involves working from observed phenomena to the generation of alternative explanations for the phenomena and comparing the alternatives to select the best possible explanation. We conclude that an ATOM can help increase the influence and impact of evidence-based prevention for population benefit.

Analyzing rid methodology through the lens of innovative abduction

Abstract: The literature reports that abduction is inherent to design reasoning. The Radical Innovation Design methodology is analyzed using the lens of Kroll and Koskela's two-step innovative abduction. In the first phase (Problem Setting), the Knowledge Design process follows a two-step selective abduction and the Problem Design process comprises abduction followed by deduction. This illustrates the specific reasoning employed when identifying the right problem on which to innovate. In the second phase (Problem Solving), the reasoning follows two-step innovative abduction.

Applying algorithm selection to abductive diagnostic reasoning

Abstract: The complexity of technical systems requires increasingly advanced fault diagnosis methods to ensure safety and reliability during operation. Particularly in domains where maintenance constitutes an extensive portion of the entire operation cost, efficient and effective failure identification holds the potential to provide large economic value. Abduction offers an intuitive concept for diagnostic reasoning relying on the notion of logical entailment. Nevertheless, abductive reasoning is an intractable problem and computing solutions for instances of reasonable size and complexity persists to pose a challenge. In this paper, we investigate algorithm selection as a mechanism to predict the “best” performing technique for a specific abduction scenario within the framework of model-based diagnosis. Based on a set of structural attributes extracted from the system models, our meta-approach trains a machine learning classifier that forecasts the most runtime efficient abduction technique given a new diagnosis problem. To assess the predictor’s selection capabilities and the suitability of the meta-approach in general, we conducted an empirical analysis featuring seven abductive reasoning approaches. The results obtained indicate that applying algorithm selection is competitive in comparison to always choosing a single abductive reasoning method.