Qualitative Process Theory

By Kenneth D. Forbus. 1984

Abstract Objects move, collide, flow, bend, heat up, cool down, stretch, compress, and boil. These and other things that cause changes in objects over time are intuitively characterized as processes . To understand commonsense physical reasoning and make programs that interact with the physical world as well as people do we must understand qualitative reasoning about processes, when they will occur, their effects, and when they will stop. Qualitative process theory defines a simple notion of physical process that appears useful as a language in which to write dynamical theories. Reasoning about processes also motivates a new qualitative representation for quantity in terms of inequalities, called the quantity space . This paper describes the basic concepts of qualitative process theory, several different kinds of reasoning that can be performed with them, and discusses its implications for causal reasoning. Several extended examples illustrate the utility of the theory, including figuring out that a boiler can blow up, that an oscillator with friction will eventually stop, and how to say that you can pull with a string, but not push with it.

Subfields: Knowledge Graphs & Representation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Knowledge Graphs & Representation

Citations in the corpus

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations (2016, background): “These efforts have grown tomodel physical processes (Forbus 1984) or tomodel a series of actions as scripts (Schank and Abelson 2013) for stories—both of which are not depicted in a single static image but which play roles in an image’s story (Vedantam et al.”
Reasoning about Quantities in Natural Language (2015, methodology): “From election results, number of casualties in accidents, to changes in stock prices, textual representations of quantities are extremely important in communicating accurate information.”
QuaRTz: An Open-Domain Dataset of Qualitative Relationship Questions (2019, background): “Despite rapid progress in general question-answering (QA), e.g., (Clark and Gardner, 2018), and formal models for qualitative reasoning (QR), e.g., (Forbus, 1984; Weld and De Kleer, 2013), there has been little work on reasoning with textual qualitative knowledge, and no dataset available in this…”
QuaRel: A Dataset and Models for Answering Questions about Qualitative Relationships (2018, methodology): “For the target formalism itself, we draw on the extensive body of work on qualitative reasoning (Forbus 1984; Weld and De Kleer 2013; Kuipers 1994) to create a logical form language that can express the required qualitative knowledge, yet is sufﬁciently constrained that parsing into it is feasible,…”
Designing Learning by Teaching Agents: The Betty's Brain System (2008, methodology): “The overall qualitative reasoning mechanism is a simplified implementation of Qualitative Process Theory (Forbus, 1984).”
Generating Explanations of Device Behavior Using Compositional Modeling and Causal Ordering (1993, background): “This is done in QP theory [8], in which causality is specified explicitly through direct influences .”
Using Qualitative Models to Guide Inductive Learning (1993, methodology): “While similar to QPT models, it should be noted that our QMs di(cid:11)er in that they are incompletely speci(cid:12)ed.”
Flowing waters or teeming crowds: Mental models of electricity (1982, background): “These qualitative-proportion relationships (see Forbus, 1982) may be phenomenological primitives, in the sense discussed by diSessa (this volume).”
The scope and limits of simulation in automated reasoning (2016, background): “• Qualitative reasoning: A collection of algorithms developed for reasoning about relative size of physical quantities and direction of change [19,36,65].”
A review of function modeling: Approaches and applications (2008, methodology): “Many of the FM developers propose to do so (Umeda et al., 1990, 1996; Umeda & Tomiyama, 1995, 1997; Chandrasakaran & Josephson, 2000; Kitamura & Mizoguchi, 2004; Yoshioka et al., 2004; Chandrasakaran, 2005), and among those, Umeda and Tomiyama implement an automated reasoning system with FM by makin”
Modeling and Simulation of Genetic Regulatory Systems: A Literature Review (2002, unknown): “Each object is described by properties that take their value from well-de (cid:142) ned domains.”
On the Representation of Physical Quantities in Natural Language Text (2004, methodology): “From our perspective, using the semantics of Qualitative Process Theory, the taxonomies suggested by Dixon and Frawley are flawed and inconsistent.”
Commonsense reasoning and commonsense knowledge in artificial intelligence (2015, background): “Later more sophisticated forms were developed in parallel by de Kleer and Brown (1985) for analyzing electronic circuits; by Forbus (1985) for analyzing varieties of physical processes; and by Kuipers (1986) as a mathematical formalism.”
Practicing versus inventing with contrasting cases: The effects of telling first on learning and transfer. (2011, background): “An invariant under transformation is a specific class of deep structure that differs from the qualitative relations (Forbus, 1984) often investigated in research on analogical transfer (e.”
Qualitative simulation of genetic regulatory networks using piecewise-linear models (2004, methodology): “The lack of quantitative information on kinetic parameters and molecular concentrations has stimulated an interest in methods for modeling and simulation developed in the field of qualitative reasoning (QR), most notably QSIM (Kuipers, 1994) and QPT (Forbus, 1984).”
Intuitive Theories (2020, background): “…reasoning and machine intelligence requires acknowledging that people's knowledge of the world is structured in terms of intuitive theories (Forbus, 1984; Gopnik & Wellman, 1992; Saxe, 2005; Wellman & Gelman, 1992), and that many cognitive functions can be understood as inferences over…”
LEARNING BY TEACHING: A NEW AGENT PARADIGM FOR EDUCATIONAL SOFTWARE (2005, unknown): “The overall qualitative reasoning mechanism is a simplified implementation of qualitative process theory (Forbus 1984).”
Representing, Running, and Revising Mental Models: A Computational Model (2018, methodology): “These are also problematic under the sole mechanism assumption (Forbus, 1984), which states that all changes in a physical system are the result of processes.”
Qualitative Spatial Representation and Reasoning: An Overview (2001, unknown): “One way to state the idea is the relevance principle: the distinctions made by a quantization must be relevant to the kind of reasoning performed [76].”