Bert Bredeweg

An Overvieuw of Approaches to Qualitative Model Construction

In qualitative reasoning research, much effort has been spent on developing representation and reasoning formalisms. Only recently, the process of constructing models in terms of these formalisms has been recognised as an important research topic of its own. Approaches addressing this topic are examined in this review. For this purpose a general model of the task of constructing qualitative models is developed that serves as a frame of reference in considering these approaches. Two categories of approaches are identified: model composition and model induction approaches. The former compose a model from predefined partial models and the latter infer a model from behavioural data. Similarities and differences between the approaches are discussed using the general task model as a reference. It appears that the majority of approaches focus on automating model construction entirely. Assessing and debugging a model in cooperation with a modeller is identified as an important topic for future research

Qualitative modeling in education

■ We argue that qualitative modeling provides a valuable way for students to learn. Two modelbuilding environments, VMODEL and HOMER/VISIGARP, are presented that support learners by constructing conceptual models of systems and their behavior using qualitative formalisms. Both environments use diagrammatic representations to facilitate knowledge articulation. Preliminary evaluations in educational settings provide support for the hypothesis that qualitative modeling tools can be valuable aids for learning.

Garp3 - Workbench for Qualitative Modelling and Simulation

Garp3 is a domain independent, multi-platform, qualitative modelling and simulation environment. It allows modellers to articulate and refine their conceptual domain knowledge and analyse this knowledge through simulation. Garp3 has been successfully applied in Ecology and Sustainable Development (SD) and is freely available via (http://www.garp3.org). Garp3 and the NaturNet-Redime Project Ecologists in the NaturNet-Redime project (http://www.naturnet.org) are building qualitative models about sustainable development issues through several case studies. For this purpose the Garp3 workbench for building, simulating, and inspecting qualitative models was developed (Bredeweg e al, 2006). The main goals of the development was making qualitative reasoning technology usable for non-computer scientists by creating a uniform user interface, a diagrammatic visual language for representing model content, and graphical buttons to communicate the available user options and manipulations. Garp3 is implemented in SWI-Prolog (http://www.swi-prolog.org) and seamlessly integrates three previously developed software components: Garp2 for simulating models, Homer for building models, and VisiGarp for inspecting simulation results. To further support the ecologists in their modelling efforts a structured approach to modelling was developed. This framework helps modellers refine their initial ideas, represented in concept maps, into detailed conceptualisations, such as structural models, causal models, and expected model behaviour. The diagrams in the final steps of the framework are close to the actual modelling primitives used to implement qualitative models. The Sketch environment in Garp3 supports modellers with tools to create the required diagrams. These Sketches not only help modellers in creating a model, but also serve as a more general description of the final model for other users of the model. One of the main goals of the NaturNet-Redime project is to create a sustainable development curriculum that allows students to learn about specific issues through qualitative modelling and simulation. To advance this goal the case study models developed in project are being integrated into a single library of sustainability concepts. Students can run and adapt different scenarios, let Garp3 automatically gather the correct knowledge relevant to the simulation and predict the possible outcomes, and analyse the results. To support the integration of the different case study models multiple model support and copy/paste functionality have been added to Garp3. The copying functionality assures that models remain syntactically correct, avoids adding redundant knowledge, preserves existing knowledge, and merges conceptual knowledge in a semantically correct manner as much as possible. The copy functionality also allows modellers to reuse parts of

Current topics in qualitative reasoning

In this editorial introduction to this special issue of AI Magazine on qualitative reasoning, we briefly discuss the main motivations and characteristics of this branch of AI research. We also summarize the contributions in this issue and point out challenges for future research.

Towards a structured approach to building qualitative reasoning models and simulations

Successful transfer and uptake of qualitative reasoning technology for modelling and simulation in a variety of domains has been hampered by the lack of a structured methodology to support formalisation of ideas. We present a framework that structures and supports the capture of conceptual knowledge about system behaviour using a qualitative reasoning approach. This framework defines a protocol for representing content that supports the development of a conceptual understanding of systems and how they behave. The framework supports modellers in two ways. First, it structures and explicates the work involved in building models. Second, it facilitates easier comparison and evaluation of intermediate and final results of modelling efforts. We show how this framework has been used in developing qualitative reasoning models about three case studies of sustainable development in different river systems.

Garp3: a new workbench for qualitative reasoning and modelling

Easy to use workbenches for Qualitative Reasoning (QR) and Modelling are virtually nonexistent. This has a limiting effect on the use and uptake of the technology by a larger audience. We present Garp3, a user-friendly workbench that allows modellers to build, simulate, and inspect qualitative models. Garp3 can be used to, discover, capture, and share conceptual knowledge on how systems behave.

A conceptual modelling framework for knowledge-level reflection

We argue for the separation of object and reflective problem solving levels and a selfrepresentation that is distinct from the object-level because it is selective, specialised and knowledge oriented, i.e., it is a knowledge-level model congruent with the KADS conceptual model of the object system. As an example we describe a conceptual model for competence assessment and improvement in Office Plan, a configuration system for office space allocation. A broad comparison with notions of reflection in logic and computational reflection clarifies the distinctiveness of our notion of knowledgelevel reflection and investigates some of the architectural options that are open for its realisation in knowledge systems. Introduction

Qualitative reasoning about population and community ecology

Traditional approaches to ecological modeling, based on mathematical equations, are hampered by the qualitative nature of ecological knowledge. In this article, we demonstrate that qualitative reasoning provides alternative and productive ways for ecologists to develop, organize, and implement models. We present a qualitative theory of population dynamics and use this theory to capture and simulate commonsense theories about population and community ecology. Advantages of this approach include the possibility of deriving relevant conclusions about ecological systems without numeric data; a compositional approach that enables the reusability of models representing partial behavior; the use of a rich vocabulary describing objects, situations, relations, and mechanisms of change; and the capability to provide causal interpretations of system behavior.

Qualitative models of interactions between two populations

Negative and positive interactions between populations such as competition, predator-prey and symbiosis, under the influence of environmental factors, have been pointed out as the main organising forces of communities. Due to explicit representation of knowledge and of causal relations, qualitative simulations can be useful for students to have insights on structure and behaviour of interacting populations. Our work is based on a library of model fragments representing basic population processes (such as natality and mortality), so that it is possible to derive complex community behaviour from ‘first principles’ [1]. Six types of interactions were modelled: neutralism (0,0), amensalism (0,−), commensalism (0,+), predator-prey (+,−), symbiosis (+,+) and competition (−,−). The symbols −, +, 0 represent, respectively, negative, positive or no effects from the other species. We define a ‘basic interaction model’, in which population1 produces some effect (effect1on2) that affects natality (born2) and mortality (dead2) of population2, while this one has an effect (effect2on1) which influences born1 and dead1. These influences may be positive or negative, according to the interaction type. For instance, in the predator-prey model effect1on2 has a positive influence on dead2 (increases mortality of prey) and effect2on1 has a negative influence on dead1 (decreases mortality of predator). Simulations with these models show typical behaviour of each interaction type, such as coexistence and competitive exclusion. The simulation models are organised according to educational learning routes, following the general scheme discussed in [2].

A qualitative reasoning model of algal bloom in the Danube Delta Biosphere Reserve (DDBR)

Abstract This paper presents a Qualitative Reasoning model of the algal bloom phenomenon and its effects in the Danube Delta Biosphere Reserve (DDBR) in Romania. Qualitative Reasoning models represent processes and their cause–effect relationships in a flexible and conceptually rich manner and as such can be used as instruments for capturing and sharing explanations of how systems behave. The DDBR model is based on expert knowledge and captures the main factors contributing to the dynamics of this aquatic ecosystem. Focal points of the model are the relationships between temperature, water pollution from the Danube catchment area, and cyanotoxins. These factors gravely affect aquatic biota and ultimately human wellbeing. In addition to capturing domain knowledge, this paper discusses solutions for representing typical patterns in ecological systems using Qualitative Reasoning techniques.