November 1995 - Vol. 38 No. 11
Features
Windows of opportunity in electronic classrooms
Paradigm-shifting landmark buildings are cherished by their occupants and remembered because they reshape our expectations of schools, homes, or offices. Classic examples include Thomas Jefferson's communal design of the “academical village” at the University of Virginia where faculty and students lived close to classrooms, Frank Lloyd Wright's organic harmony with nature in Fallingwater (in western Pennsylvania) where the waterfall sounds and leafy surroundings offered a stress-reducing getaway for an urban executive, or Kevin Roche's open glass-walled Ford Foundation (in New York City) that promoted new team-oriented management strategies.
Welcome to the first installment of Digital Village. This new column will be a reliable source of information on modern digital network technologies, particularly from the client side, and the use of those technologies for the betterment of society. In an attempt to provide information on cyberspace and its tools, we hope to be useful to readers in maintaining currency and perspective.
New horizons in commercial and industrial AI
AI as a field has undergone rapid growth in diversification and practicality. For the past 10 years, the repertoire of AI techniques has evolved and expanded. Scores of newer fields have recently been added to the traditional domains of practical AI. Although much practical AI is still best characterized as advanced computing rather than intelligence, applications in everyday commercial and industrial settings have certainly increased, especially since 1990. Additionally, AI has shown a growing influence on other computer science areas, such as databases, software engineering, distributed computing, computer graphics, user interfaces, and stimulation.
CYC: a large-scale investment in knowledge infrastructure
Since 1984, a person-century of effort has gone into building CYC, a universal schema of roughly 105 general concepts spanning human reality. Most of the time has been spent codifying knowledge about these concepts; approximately 106 commonsense axioms have been handcrafted for and entered into CYC's knowledge base, and millions more have been inferred and cached by CYC. This article examines the fundamental assumptions of doing such a large-scale project, reviews the technical lessons learned by the developers, and surveys the range of applications that are or soon will be enabled by the technology.
WordNet: a lexical database for English
Because meaningful sentences are composed of meaningful words, any system that hopes to process natural languages as people do must have information about words and their meanings. This information is traditionally provided through dictionaries, and machine-readable dictionaries are now widely available. But dictionary entries evolved for the convenience of human readers, not for machines. WordNet1 provides a more effective combination of traditional lexicographic information and modern computing. WordNet is an online lexical database designed for use under program control. English nouns, verbs, adjectives, and adverbs are organized into sets of synonyms, each representing a lexicalized concept. Semantic relations link the synonym sets [4].
Natural language processing will grow into a vital industrial technology in the next five to 10 years. But this growth depends on the development of large linguistic databases that capture natural language phenomena [1, 2]. Another important theme for future work is development of large knowledge bases that are shared widely by different groups. One promising approach to such knowledge bases draws on natural language processing and linguistic knowledge. This article describes the EDR Electronic Dictionary [3], which seeks to provide a foundation for linguistic databases, and explains the relation of electronic dictionaries to very large knowledge bases.
CYC, WordNet, and EDR: critiques and responses
I applaud Miller's WordNet project and feel that there is much in common in our approaches, even though there are fundamental differences in the two expressions of that spirit. Here, I list the four differences I noted, closing with a crucial observation concerning the common spirit in our work.
Industrial applications of distributed AI
Most work done in distributed artificial intelligence (DAI) had targeted sensory networks, including air traffic control, urban traffic control, and robotic systems. The main reason is that these applications necessitate distributed interpretation and distributed planning by means of intelligent sensors. Planning includes not only the activities to be undertaken, but also the use of material and cognitive resources to accomplish interpretation tasks and planning tasks. These application areas are also characterized by a natural distribution of sensors and receivers in space. In other words, the sensory data-interpretation tasks and action planning are inter-dependent in time and space. For example, in air traffic control, a plan for guiding an aircraft must be coordinated with the plans of other nearby aircraft to avoid collisions.
Applications of machine learning and rule induction
Machine learning is the study of computational methods for improving performance by mechanizing the acquisition of knowledge from experience. Expert performance requires much domain-specific knowledge, and knowledge engineering has produced hundreds of AI expert systems that are now used regularly in industry. Machine learning aims to provide increasing levels of automation in the knowledge engineering process, replacing much time-consuming human activity with automatic techniques that improve accuracy or efficiency by discovering and exploiting regularities in training data. The ultimate test of machine learning is its ability to produce systems that are used regularly in industry, education, and elsewhere.
Applications of inductive logic programming
Techniques of machine learning have been successfully applied to various problems [1, 12]. Most of these applications rely on attribute-based learning, exemplified by the induction of decision trees as in the program C4.5 [20]. Broadly speaking, attribute-based learning also includes such approaches to learning as neural networks and nearest neighbor techniques. The advantages of attribute-based learning are: relative simplicity, efficiency, and existence of effective techniques for handling noisy data. However, attribute-based learning is limited to non-relational descriptions of objects in the sense that the learned descriptions do not specify relations among the objects' parts. Attribute-based learning thus has two strong limitations:
the background knowledge can be expressed in rather limited form, and
the lack of relations makes the concept description language inappropriate for some domains.
Commercial applications of natural language processing
Vast quantities of text are becoming available in electronic form, ranging from published documents (e.g., electronic dictionaries, encyclopedias, libraries and archives for information retrieval services), to private databases (e.g., marketing information, legal records, medical histories), to personal email and faxes. Online information services are reaching mainstream computer users. There were over 15 million Internet users in 1993, and projections are for 30 million in 1997. With media attention reaching all-time highs, hardly a day goes by without a new article on the National Information Infrastructure, digital libraries, networked services, digital convergence or intelligent agents. This attention is moving natural language processing along the critical path for all kinds of novel applications.
Object-oriented AI: a commercial perspective
Computer journals and magazines are currently filled with stories about object technology, just as they were filled with stories about artificial intelligence (AI) and expert systems in the early 1980s. The naive reader might think that expert systems technology had failed and that object-oriented technology was the hot new software approach of the 1990s.
Rough set theory, introduced by Zdzislaw Pawlak in the early 1980s [11, 12], is a new mathematical tool to deal with vagueness and uncertainty. This approach seems to be of fundamental importance to artificial intelligence (AI) and cognitive sciences, especially in the areas of machine learning, knowledge acquisition, decision analysis, knowledge discovery from databases, expert systems, decision support systems, inductive reasoning, and pattern recognition.
Principles and applications of chaotic systems
There lies a behavior between rigid regularity and randomness based on pure chance. It's called a chaotic system, or chaos for short [5]. Chaos is all around us. Our notions of physical motion or dynamic systems have encompassed the precise clock-like ticking of periodic systems and the vagaries of dice-throwing chance, but have often been overlooked as a way to account for the more commonly observed chaotic behavior between these two extremes. When we see irregularity we cling to randomness and disorder for explanations. Why should this be so? Why is it that when the ubiquitous irregularity of engineering, physical, biological, and other systems are studied, it is assumed to be random and the whole vast machinery of probability and statistics is applied? Rather recently, however, we have begun to realize that the tools of chaos theory can be applied toward the understanding, manipulation, and control of a variety of systems, with many of the practical applications coming after 1990. To understand why this is true, one must start with a working knowledge of how chaotic systems behave—profoundly, but sometimes subtly different, from the behavior of random systems.
Since deterministic chaos is not only a profound concept in science but also a ubiquitous phenomenon in real-world nonlinear systems, extending to a variety of temporal and spatial scales, it can be naturally related to applications in science and technology [4]. In fact, it is not difficult to find the buds of such possible applications in historical papers by Van der Pol and Van der Mark [22], Ulam and von Neumann [21], and Kalman [12], although the term deterministic chaos was not used in those days.
Artificial life meets entertainment: lifelike autonomous agents
The relatively new field of artificial life attempts to study and understand biological life by synthesizing artificial life forms. To paraphrase Chris Langton, the founder of the field, the goal of artificial life is to “model life as it could be so as to understand life as we know it.” Artificial life is a very broad discipline which spans such diverse topics as artificial evolution, artificial ecosystems, artificial morphogenesis, molecular evolution, and many more. Langton offers a nice overview of the different research questions studied by the discipline [6]. Artificial life shares with artificial intelligence (AI) its interest in synthesizing adaptive autonomous agents. Autonomous agents are computational systems that inhabit some complex, dynamic environment, sense and act autonomously in this environment, and by doing so realize a set of goals or tasks for which they are designed.
Australian attitudes toward legal intervention into hacking
The Australian public seems to have a characteristic response to computer crime. It is worth giving some anecdotal evidence to support this statement. During fieldwork [1-5], respondents were quizzed regarding their reasoning for their responses to the criminality of hacking. Some results are given here in response to the social attitudes of Australians towards reporting evidence of the enactment of computer crimes (see Table 1). This is in support of Thompson's [12, 13] and others' claims that relatively few computer crimes are, indeed, reported to the police in Australia (see also [9]).
When computers are used to control potentially dangerous devices, new issues and concerns are raised for software engineering. Simply focusing on building software that matches its specifications is not enough. Accidents occur even when the individual system components are highly reliable and have not “failed.” That is, accidents in complex systems often arise in the interactions among the system components, each one operating according to its specified behavior but together creating a hazardous system state.
