Representation Of Structured Objects Research Articles

AI—A multiple book review

Textbooks on Artificial Intelligence offer a unique trace of the development of our discipline. The topics and the focus have been changing over the decades: The first AI book I put my hands on was [4] and astonishing enough, the latest AI book I received is from the same author, [6]. Nearly 30 years of research have led to a considerable difference between the two books. While the first one was dealing, as the title promises, with “Problem Solving Methods in Artificial Intelligence”, the latter is aiming at the introduction of a new synthesis, a new definition of our field. In the 1971 book (and in the 1980 book [5] on “Principles of Artificial Intelligence”) the main sections cover search, resolution, plan generation and structured object representation, i.e., a careful development of semantic networks starting from a predicate logic description. This aspect became more important with the success of knowledge based systems. And indeed the eighties, the high time of knowledge representation, led to AI books focusing on various knowledge representation formats. In [8] and [2] considerable parts deal with knowledge representation formats like frames, conceptual dependencies or semantic networks. There was a clear shift from general problem solving techniques, like heuristic search and inference, towards knowledge representation and its applications. The eighties witnessed the flourishing of AI—knowledge based systems and expert systems seemed to be the main contribution of AI to the rest of computer science and hence the various formats for representing knowledge have been considered to play an essential role. In the nineties this situation changed drastically: it turned out that a deep understanding of the various representation languages is only possible if they are mapped onto logical formalisms. But even more, it became obvious that logical semantics of semantic networks, frames and similar formats are the key to the analysis and design of efficient inference systems. As a consequence specialized logical systems became the main ingredients of the AI-toolbox. During the eighties, however, not only did we experience the high time of knowledge based system, at the same time “Nouvelle AI” entered the floor: bottom up, situated, embodied and following the premise that “Intelligence is in the eye of the observer” [1]. Nowadays, numerous groups are researching probabilistic approaches, neural networks or

Comparing Structures Using a Hopfield-Style Neural Network

Labeled graphs are an appropriate and popular representation of structured objects in many domains. If the labels describe the properties of real world objects and their relations, finding the best match between two graphs turns out to be the weakly defined, NP-complete task of establishing a mapping between them that maps similar parts onto each other preserving as much as possible of their overall structural correspondence. In this paper, former approaches of structural matching and constraint relaxation by spreading activation in neural networks and the method of solving optimization tasks using Hopfield-style nets are combined. The approximate matching task is reformulated as the minimization of a quadratic energy function. The design of the approach enables the user to change the parameters and the dynamics of the net so that knowledge about matching preferences is included easily and transparently. In the last section, some examples demonstrate the successful application of this approach in classification and learning in the domain of organic chemistry.

Recent Advances in Graph Matching

A powerful and universal data structure with applications invarious subfields of science and engineering is graphs. In computer vision and image analysis, graphs are often used for the representation of structured objects. For example, if the problem is to recognize instances of known objects in an image, then often models, or prototypes, of the known objects are represented by means of graphs and stored in a database. The unknown objects in the input image are extracted by means of suitable preprocessing and segmentation algorithms, and represented by graphs that are analogous to the model graphs. Thus, the problem of object recognition is transformed into a graph matching problem. In this paper, it is assumed that there is an input graph that is given on-line, and a number of model, or prototype, graphs that are known a priori. We present a new approach to subgraph isomorphism detection which is based on a compact representation of the model graphs that is computed off-line. Subgraphs that appear multiple times within the same or within different model graphs are represented only once, thus reducing the computational effort to detect them in an input graph. In the extreme case where all model graphs are highly similar, the run time of the new algorithm becomes independent of the number of model graphs. We also describe an extension of this method to error-correcting graph matching. Furthermore, an approach to subgraph isomorphism detection based on decision trees is proposed. A decision tree is generated from the models in an off-line phase. This decision tree can be used for subgraph isomorphism detection. The time needed for decision tree traversal is only polynomial in terms of the number of nodes of the input graph. Moreover, the time complexity of the decision tree traversal is completely independent on the number of model graphs, regardless of their similarity. However, the size of the decision tree is exponential in the number of nodes of the models. To cut down the space complexity of the decision tree, some pruning strategies are discussed.

Intelligent simulation for flexible manufacturing systems: An integrated approach

The use of simulation in the design and control of manufacturing systems has become increasingly popular in the last few years. Although simulation has proven to be superior to traditional mathematical modeling in capturing the scope, complexity and dynamics of manufacturing systems, it remains merely a descriptive modeling tool. This leaves the user with the difficult tasks of generating, testing, and selecting design and operation alternatives. In this paper, a methodology that describes the supplementation of simulation systems with these more intelligent and prescriptive functions is proposed. A new architecture, based on a structured object representation (SOR) paradigm, is suggested. It is shown that the adoption of an SOR paradigm results in easier and more effective low level integration of the modeling, simulation and reasoning functions. The design of an experimental prototype system that allows automatic model generation and simulation as well as model evaluation is also presented.

Knowledge acquisition using hypertext

This paper addresses the issue of supporting knowledge acquisition using hypertext. We propose a way of tightly integrating hypertext and structured object representation , using Artificial Intelligence (AI) frames for the basic representation of hypertext nodes. Epistemologically, a dual view of the resulting space is of interest. One view is that of hypertext which emphasizes nodes containg g text, including formal knowledge representation. The other view focuses on objects with certain relationships, which define a semantic network . Both in hypertext and in semantic networks the relations between chunks of knowledge are explicitly represented by links. However, in today's hypertext systems a node typically contains just informal text and references to other nodes. Our approach additionally facilitates the explicit representation of structure “inside” hypertext nodes using partitions . We show the usefulness of such a tight integration for knowledge acquisition, providing several features useful for supporting it based on a level of basic hypertext functionality. In particular, we sketch a method for doing knowledge acquisition in such an environment. Hypertext is used as a mediating “semiformal” representation, which allows experts to directly represent knowledge without the immediate support of knowledge engineers. These help then to make this knowledge operational, supported by the system's facility to provide templates as well as their links to the semiformal representation. As an example of our results of using this method of knowledge acquisition, we illustrate the strategic knowledge in our application domain. More generally, our approach supports important aspects of (software) engineering knowledge-based systems and their maintenance. Also their user interface can be improved this way.

Structured Object Representations in Many-Sorted Attribute Systems

Units (frames) and semantic networks are the most popular schemes for representation of structured objects. Both schemes can be formaly interpreted in a predicate calculus. In the paper a new representational scheme, based on so-called many-sorted attribute systems, is proposed. We define elementary attributes and four types of structured attributes (corresponding to four types of structuring: union - subclasses of a class of objects, meet - properties of objects of the same class, product - subobjects of an object, set - a collection of elements of an ob ject). With this attributes we can describe a structured object, consisting of some simpler objects. A description of an object is given directly in a structured attribute defined by its component attributes. For this purpose, we associate a formalized language with a collection of attributes, defining syntax, semantics of this language. By analogy to semantic networks we propose so-called attribute networks and inversely to units so-called reversed units. Interpretation of this new structured object representations follows from the many-sorted attribute systems presented in the paper.