• 제목/요약/키워드: causal inference

Search Result 68, Processing Time 0.027 seconds

Category-Based Feature Inference: Testing Causal Strength (범주기반 속성추론: 인과관계 강도의 검증)

  • JunHyoung Jo;Hyung-Chul O. Li;ShinWoo Kim
    • Science of Emotion and Sensibility
    • /
    • v.26 no.1
    • /
    • pp.55-64
    • /
    • 2023
  • This research investigated category-based feature inference when category features were connected in common cause and common effect causal networks. Previous studies that tested feature inference in causal categories showed unique inference patterns depending on causal direction, number of related features, whether the to-be-inferred feature was cause or effect, etc. However, these prior studies primarily focused on inference pattens that arise from causal relations, and few studies directly explored how the effects of causal relations vary depending on causal strength. We tested feature inference in common cause (Expt. 1) and common effect (Expt. 2) causal categories when casual strengths were either strong or weak. To this end, we had participants learn causal categories where features were causally linked and then perform feature inference task. The results showed that causal strengths as well as causal relations had important impacts on feature inference. When causal strength was strong, inference for common cause feature became weaker but that for the common effect feature became stronger. Moreover, when causal strength was strong and common cause was present, inference for the effect features became stronger, whereas the results were reversed in common effect networks. In particular, in common effect networks, casual discounting was more evident with strong causal strength. These results consistently demonstrate that participants consider not only causal relations but also causal strength in feature inference of causal categories.

Causality, causal discovery, causal inference and counterfactuals in Civil Engineering: Causal machine learning and case studies for knowledge discovery

  • M.Z. Naser;Arash Teymori Gharah Tapeh
    • Computers and Concrete
    • /
    • v.31 no.4
    • /
    • pp.277-292
    • /
    • 2023
  • Much of our experiments are designed to uncover the cause(s) and effect(s) behind a phenomenon (i.e., data generating mechanism) we happen to be interested in. Uncovering such relationships allows us to identify the true workings of a phenomenon and, most importantly, to realize and articulate a model to explore the phenomenon on hand and/or allow us to predict it accurately. Fundamentally, such models are likely to be derived via a causal approach (as opposed to an observational or empirical mean). In this approach, causal discovery is required to create a causal model, which can then be applied to infer the influence of interventions, and answer any hypothetical questions (i.e., in the form of What ifs? Etc.) that commonly used prediction- and statistical-based models may not be able to address. From this lens, this paper builds a case for causal discovery and causal inference and contrasts that against common machine learning approaches - all from a civil and structural engineering perspective. More specifically, this paper outlines the key principles of causality and the most commonly used algorithms and packages for causal discovery and causal inference. Finally, this paper also presents a series of examples and case studies of how causal concepts can be adopted for our domain.

Category-based Feature Inference in Causal Chain (인과적 사슬구조에서의 범주기반 속성추론)

  • Choi, InBeom;Li, Hyung-Chul O.;Kim, ShinWoo
    • Science of Emotion and Sensibility
    • /
    • v.24 no.1
    • /
    • pp.59-72
    • /
    • 2021
  • Concepts and categories offer the basis for inference pertaining to unobserved features. Prior research on category-based induction that used blank properties has suggested that similarity between categories and features explains feature inference (Rips, 1975; Osherson et al., 1990). However, it was shown by later research that prior knowledge had a large influence on category-based inference and cases were reported where similarity effects completely disappeared. Thus, this study tested category-based feature inference when features are connected in a causal chain and proposed a feature inference model that predicts participants' inference ratings. Each participant learned a category with four features connected in a causal chain and then performed feature inference tasks for an unobserved feature in various exemplars of the category. The results revealed nonindependence, that is, the features not only linked directly to the target feature but also to those screened-off by other feature nodes and affected feature inference (a violation of the causal Markov condition). Feature inference model of causal model theory (Sloman, 2005) explained nonindependence by predicting the effects of directly linked features and indirectly related features. Indirect features equally affected participants' inference regardless of causal distance, and the model predicted smaller effects regarding causally distant features.

A Causal Knowledge-Driven Inference Engine for Expert System

  • Lee, Kun-Chang;Kim, Hyun-Soo
    • Journal of the Korean Institute of Intelligent Systems
    • /
    • v.8 no.6
    • /
    • pp.70-77
    • /
    • 1998
  • Although many methods of knowledge acquisition has been developed in the exper systems field, such a need form causal knowledge acquisition hs not been stressed relatively. In this respect, this paper is aimed at suggesting a causal knowledge acquisition process, and then investigate the causal knowledge-based inference process. A vehicle for causal knowledge acquisition is FCM (Fuzzy Cognitive Map), a fuzzy signed digraph with causal relationships between concept variables found in a specific application domain. Although FCM has a plenty of generic properties for causal knowledge acquisition, it needs some theoretical improvement for acquiring a more refined causal knowledge. In this sense, we refine fuzzy implications of FCM by proposing fuzzy causal relationship and fuzzy partially causal relationship. To test the validity of our proposed approach, we prototyped a causal knowledge-driven inference engine named CAKES and then experimented with some illustrative examples.

  • PDF

Causal inference from nonrandomized data: key concepts and recent trends (비실험 자료로부터의 인과 추론: 핵심 개념과 최근 동향)

  • Choi, Young-Geun;Yu, Donghyeon
    • The Korean Journal of Applied Statistics
    • /
    • v.32 no.2
    • /
    • pp.173-185
    • /
    • 2019
  • Causal questions are prevalent in scientific research, for example, how effective a treatment was for preventing an infectious disease, how much a policy increased utility, or which advertisement would give the highest click rate for a given customer. Causal inference theory in statistics interprets those questions as inferring the effect of a given intervention (treatment or policy) in the data generating process. Causal inference has been used in medicine, public health, and economics; in addition, it has received recent attention as a tool for data-driven decision making processes. Many recent datasets are observational, rather than experimental, which makes the causal inference theory more complex. This review introduces key concepts and recent trends of statistical causal inference in observational studies. We first introduce the Neyman-Rubin's potential outcome framework to formularize from causal questions to average treatment effects as well as discuss popular methods to estimate treatment effects such as propensity score approaches and regression approaches. For recent trends, we briefly discuss (1) conditional (heterogeneous) treatment effects and machine learning-based approaches, (2) curse of dimensionality on the estimation of treatment effect and its remedies, and (3) Pearl's structural causal model to deal with more complex causal relationships and its connection to the Neyman-Rubin's potential outcome model.

Modeling feature inference in causal categories (인과적 범주의 속성추론 모델링)

  • Kim, ShinWoo;Li, Hyung-Chul O.
    • Korean Journal of Cognitive Science
    • /
    • v.28 no.4
    • /
    • pp.329-347
    • /
    • 2017
  • Early research into category-based feature inference reported various phenomena in human thinking including typicality, diversity, similarity effects, etc. Later research discovered that participants' prior knowledge has an extensive influence on these sorts of reasoning. The current research tested the effects of causal knowledge on feature inference and conducted modeling on the results. Participants performed feature inference for categories consisted of four features where the features were connected either in common cause or common effect structure. The results showed typicality effects along with violations of causal Markov condition in common cause structure and causal discounting in common effect structure. To model the results, it was assumed that participants perform feature inference based on the difference between the probabilities of an exemplar with the target feature and an exemplar without the target feature (that is, $p(E_{F(X)}{\mid}Cat)-p(E_{F({\sim}X)}{\mid}Cat)$). Exemplar probabilities were computed based on causal model theory (Rehder, 2003) and applied to inference for target features. The results showed that the model predicts not only typicality effects but also violations of causal Markov condition and causal discounting observed in participants' data.

Fuzzy Causal Knowledge-Based Expert System

  • Lee, Kun-Chang;Kim, Hyun-Soo;Song, Yong-Uk
    • Proceedings of the Korean Institute of Intelligent Systems Conference
    • /
    • 1998.10a
    • /
    • pp.461-467
    • /
    • 1998
  • Although many methods of knowledge acquisition has been developed in the expert systems field, such a need for causal knowledge acquisition has not been stressed relatively. In this respect, this paper is aimed at suggesting a causal knowledge acquisition process, and then investigate the causal knowledge-based inference process. A vehicle for causal knowledge acquisition is FCM (Fuzzy Cognitive Map), a fuzzy signed digraph with causal relationships between concept variables found in a specific application domain. Although FCM has a plenty of generic properties for causal knowledge acquisition, it needs some theoretical improvement for acquiring a more refined causal knowledge. In this sense, we refine fuzzy implications of FCM by proposing fuzzy implications of FCM by proposing fuzzy causal relationship and fuzzy partially causal relationship. To test the validity of our proposed approcach, we prototyped a causal knowledge-driven inference engine named CAKES and then experime ted with some illustrative examples.

  • PDF

Practice of causal inference with the propensity of being zero or one: assessing the effect of arbitrary cutoffs of propensity scores

  • Kang, Joseph;Chan, Wendy;Kim, Mi-Ok;Steiner, Peter M.
    • Communications for Statistical Applications and Methods
    • /
    • v.23 no.1
    • /
    • pp.1-20
    • /
    • 2016
  • Causal inference methodologies have been developed for the past decade to estimate the unconfounded effect of an exposure under several key assumptions. These assumptions include, but are not limited to, the stable unit treatment value assumption, the strong ignorability of treatment assignment assumption, and the assumption that propensity scores be bounded away from zero and one (the positivity assumption). Of these assumptions, the first two have received much attention in the literature. Yet the positivity assumption has been recently discussed in only a few papers. Propensity scores of zero or one are indicative of deterministic exposure so that causal effects cannot be defined for these subjects. Therefore, these subjects need to be removed because no comparable comparison groups can be found for such subjects. In this paper, using currently available causal inference methods, we evaluate the effect of arbitrary cutoffs in the distribution of propensity scores and the impact of those decisions on bias and efficiency. We propose a tree-based method that performs well in terms of bias reduction when the definition of positivity is based on a single confounder. This tree-based method can be easily implemented using the statistical software program, R. R code for the studies is available online.

The Students' Causal Inference Modes on Experimental Evidence Evaluation for Optical Phenomena (광학 현상 증거 해석의 인과적 추론 방식)

  • Pak, Sung-Jae;Jang, Byung-Ghi
    • Journal of The Korean Association For Science Education
    • /
    • v.14 no.2
    • /
    • pp.123-132
    • /
    • 1994
  • The experimental evidence evaluation of the 11th grade students(N:91) was investigated. Specially, the influence of students' ideas about optical phenomena and presented evidence types on their evidence evaluation, and the influence of students' ideas on their causal inference modes were investigated. After eliciting the students' ideas about shadow phenomena and conformity of their idea, the experimental results with a binary outcome were presented as the evidence. Then the students were asked to evaluate the evidence. Again students' ideas were elicited. Most of students had causal ideas such that the shape of object(96%) and the inclination of screen(75%) were causes of shadow shape, not the shape(70%) and color(92%) of light source. In the case of the shape of object and the color of light source, most students(70%) believed strongly their ideas. Most responses(80%) in the evidence were evidence-based, and 12% of them were theory-based. There was no significant difference of reponses types between students with causal ideas(81%) and students with non-causal ideas(78%), between covariable and non-covariable evidence. But in the case of non-causal ideas, covariable evidence was more likely to yield evidence-based reponses than non-covariable evidence. If students had preconcepts inconsistent(84%) with the evidence, they were more likely to make evidence-based responses than the students with consistent ideas (75%) with the evidence. Especially in the case perceptually biased evidence, this tendency was marked. In the case of covariable evidence, many students made inclusion inferences(40%) rather than uncertainty inferences(32%). In the case of uncertainty inferences(94%), students more likely to make evidence-based reponses than inclusion inferences(83%) and exclusion infernces(88%). In the case of inclusion inferences and exclusion infernces, students tended to make idea-based responses and distort the evidences. In conclusion, when the students evaluate the experimental evidences, their ideas influence the causal inference modes. Especially, according to the conformity of the preconcepts and logical relation of evidences, the inference modes are more strongly depended upon the preconcepts rather than evidences.

  • PDF

Development of Expert System for Diagnosis of Weld Defects (용접 결함 진단 전문가시스템의 개발)

  • 박주용
    • Journal of Advanced Marine Engineering and Technology
    • /
    • v.20 no.1
    • /
    • pp.13-23
    • /
    • 1996
  • Weld defects degrade the strength and safety of astructure and are resulted from the various cases. The complexity of causal relation of weld defects requires an expert for the analysis of weld defects and the measures counter to them. An expert system has the intelligent functions such as the representation of knowledge and the inference. On this research, weld defect are systematically analysed and their causal model is developed. This information is saved to the knowledge base. The suitable inference algorithm for the diagnosis of weld defects is developed and realized with C++ programming.

  • PDF