To improve the interaction between students and an intelligent tutoring system, we developed two Natural Language generators, that we systematically evaluated in a three way comparison that included the original system as well. We found that the generator which intuitively produces the best language does engender the most learning. Specifically, it appears that functional aggregation is responsible for the improvement. 1

Contextual vocabulary acquisition (CVA) is the active, deliberate acquisition of a meaning for an unknown word in a text by reasoning from textual clues, prior knowledge, and hypotheses developed from prior encounters with the word, but without external sources of help such as dictionaries or people. Published strategies for doing CVA vaguely and unhelpfully tell the reader to “guess”. AI algorithms for CVA can fill in the details that replace “guessing ” by “computing”; these details can then be converted to a curriculum that can be taught to students to improve their reading comprehension. Such algorithms also suggest a way out of the Chinese Room and show how holistic semantics can withstand certain objections. 1 1 Computational Philosophy and Philosophical Computation Computer science in general, and AI in particular, have a lot to give to philosophy, and vice versa, as Daniel Dennett once noted (1978: 126; cf. Rapaport 1986b). This essay discusses an interdisciplinary, applied cognitive-science research project that exhibits how philosophy can influence AI, how AI can influence philosophy, and how both can influence educational practice. I 1 take “computational philosophy ” to be the application of computational (i.e., algorithmic) solutions to philosophical problems. 2 An example from my own research would be the use of the SNePS knowledgerepresentation,

of an embodied computational cognitive agent called Cassie, based on the Grounded Layered Architecture with Integrated Reasoning (GLAIR). In this document we describe a new implementation, in which Cassie's body and the world are simulated in Crystal Space, an environment for building 3-D games. We describe the implementation of Cassie in a Crystal Space environment including her current suite of actions and her simulated vision system. Crystal Cassie is a tool for cognitive modeling and testing cognitive theories.

Abstract not found

Abstract. Brittleness is a common problem among AI systems. Autonomous systems, including those that learn, may be faced with unanticipated situations that cause decreased performance, or in the worstcase, catastrophic failures from which the system cannot recover. In this paper, we describe a construct called the metacognitive loop (MCL) that allows AI systems to monitor their own behavior, generate expectations about their own progress and performance, and verify that they are met. When expectations are violated, the metacognitive loop attempts to reason in a domain-general way about why expectations were not met and how to recover. The basis for reasoning is a set of ontologies that encode abstract diagnosic and prescriptive processes for coping with failures. 1

Abstract not found

The standard development of a dialogue system today involves the following steps: corpus collection and analysis, system development guided by corpus analysis, and finally, rigorous evaluation. Often, evaluation may involve more than one version of the system, for example when it is desirable to show the effect of system parameters that differ from one version to another.

Abstract. This note describes a logical system based on concepts and contexts, whose aim is to serve as a representation language for meanings of natural language sentences. The logic is a theoretical description of the output of an evolving implemented system, the system Bridge, which we are developing at parc, as part of the aquaint program. The note concentrates on the results of an experiment which changed the underlying ontology of the representation language from cyc to a version of WordNet/VerbNet. 1

GLAIR (Grounded Layered Architecture with Integrated Reasoning) is a multi-layered cognitive architecture for embodied agents operating in real, virtual, or simulated environments containing other agents. The highest layer of the GLAIR Architecture, the Knowledge Layer (KL), contains the beliefs of the agent, and is the layer in which conscious reasoning, planning, and act selection is performed. The lowest layer of the GLAIR Architecture, the Sensori-Actuator Layer (SAL), contains the controllers of the sensors and effectors of the hardware or software robot. Between the KL and the SAL is the Perceptuo-Motor Layer (PML), which grounds the KL symbols in perceptual structures and subconscious actions, contains various registers for providing the agent’s sense of situatedness in the environment, and handles translation and communication between the KL and the SAL. The motivation for the development of GLAIR has been “Computational Philosophy”, the computational understanding and implementation of human-level intelligent behavior without necessarily being bound by the actual implementation of the human mind. Nevertheless, the approach has been inspired by human psychology and biology.

The goal of our work is to improve the Natural Language feedback provided by Intelligent Tutoring Systems. In this paper, we discuss how to make the content presented by one such system more ¤uent and comprehensible, and we show how we accomplish this by using relatively inexpensive domain-independent text structuring techniques. We show how speci£c rhetorical relations can be introduced based on the data itself in a bottom-up fashion rather than being planned top-down by the discourse planner.