A Text-Based Intelligent System should provide more in-depth information about the contents of its corpus than does a standard information retrieval system, while at the same time avoiding the complexity and resource-consuming behavior of detailed text understanders. Instead of focusing on discovering documents that pertain to some topic of interest to the user, an approach is introduced based on the criterion of directionality (e.g., Is the agent in favor of, neutral, or opposed to the event?). A method is described for coercing sentence meanings into a metaphoric model such that the only semantic interpretation needed in order to determine the directionality of a sentence is done with respect to the model. This interpretation method is designed to be an integrated component of a hybrid information access system. 1 Introduction In the light of the increasing availability of computer-accessible full text, an important goal of a Text-Based Intelligent System is to provide a me...

Over the past ten years, researchers studying the structure of discourse have consistently had to face questions such as the following: Given that discourses consist of segments, how do the segments relate? What intersegment relations are there? How many are needed? A fair amount of controversy exists, ranging from the parsimonious position (that two basic relations suffice) to the profligate position (that an open-ended set of semantic/rhetorical relations is required). This paper outlines the arguments and then summarizes a survey of the conclusions of approximately 30 researchers -- from linguists to computational linguists to philosophers to Artificial Intelligence workers. It fuses and taxonomizes the more than 400 relations they have proposed into a hierarchy of approximately 70 increasingly semantic relations, and argues that though the taxonomy is open-ended in one dimension, it is bounded in the other and therefore does not give rise to anarchy. Some evidence is provided for the organization of the taxonomy, aswell as a full listing of the sources.

This paper consists of two parts. The first part discusses commonsense knowledge about events as manifested in language. Three kinds of knowledge are identified: compositional, durational, and aspectual. Compositional knowledge concerns internal structuring of events into preparatory, initial, main (the body), final, and resulting stages. Durational knowledge concerns durational relations between events and stages of the same event. Durational knowledge can be expressed as qualitative dependencies among the parameters of the event and as its time scale. The notion of time scale is introduced and related to shared cyclical events (time units). In discussing

This paper describes a natural language system which improves its own performance through learning. The system processes short English narratives and is able to acquire, from a single narrative, a new schema for a stereotypical set of actions. During the understanding process, the system attempts to construct explanations for characters ' actions in terms of the goals their actions were meant to achieve. When the system observes that a character has achieved an interesting goal in a novel way, it generalizes the set of actions they used to achieve this goal into a new schema. The generalization process is a knowledge-based analysis of the causal structure of the narrative which removes unnecessary details while maintaining the validity of the causal explanation. The resulting generalized set of actions is then stored as a new schema and used by the system to correctly process narratives which were previously beyond its capabilities. I

In order to build a new breed of software that can deeply understand people and our problems, so that they can help us to solve them, we are developing at the Media Lab a suite of computational tools to give machines the capacity to learn and reason about everyday life---in other words, to give machines `common sense'. We are building several large-scale commonsense knowledge bases that model broad aspects of the ordinary human world, including descriptions of the kinds of goals people have, the actions we can take and their effects, the kinds of objects that we encounter every day, and so forth, as well as the relationships between such entities. In this article we describe three systems we have built---ConceptNet, LifeNet, and StoryNet---that take unconventional approaches to representing, acquiring, and reasoning with large quantities of commonsense knowledge. Each adopts a different approach: ConceptNet is a large-scale semantic network, LifeNet is a probabilistic graphical model, and StoryNet is a database of story-scripts. We describe the evolution of these three systems, the techniques that underlie their construction and their operation, and conclude with a discussion of how we might combine them into an integrated commonsense reasoning system that uses multiple representations and reasoning methods.

The provision of summaries is of crucial importance for fully effective retrieval of information, but research on summarization has been relatively neglected, After an outline of the basic linguistic and cognitive complexities of text understanding and summarizing, the paper reviews some current projects towards automating various aspects of summarization, and discusses future prospects. 1.

Humans naturally share knowledge by telling stories. This is a form of knowledge exchange we engage in right from early childhood, and over time we learn to recall, order and organize our experiences as stories [1]. In this paper we describe the Open Mind Experience(OMEX) system, a web-based knowledge acquisition tool that exploits our natural ability to tell and explain stories in order to build a large-scale commonsense knowledgebase. We built OMEX to gather descriptions and explanations of everyday, 'common sense' experiences from volunteer contribu- tors distributed across the Intemet. We first describe the knowledge from the general public, the Open Mind Common Sense (OMCS) project. The OMCS project focused on collecting largely assertional commonsense knowledge, and we describe some of its products and spin-offs. We then give several motivating reasons for why we now wish to now collect more script-like knowledge. We then explain the features of the new OMEX site and give an evaluation of system based on a preliminary user study. We conclude by discussing our future directions.

Abstract: Business Process Re-engineering (BPR) is an area that requires a lot of technical documents and an important feature of a well-written document is a coherent narrative. Even though computer software has helped authors in many other aspects of writing, support for document narratives is almost non-existent. Therefore, we introduce CANS (Computer-Aided Narrative Support), a tool that uses Rhetorical Structure Theory to enhance the narrative of a document. From this narrative, the tool generates questions to prompt the author for the content of the document. CANS also allows the author to explore alternative narratives for a document. A catalogue of predefined narrative structures for popular types of documents is provided too. Our tool is still in its rudimentary stages but sufficiently complete to be demonstrated. 1

Stories can encapsulate complexity, subtlety, and nuance: all of which are implicitly contained in narrative and reasoned about automatically through the mental processes that come naturally to humans. For example, humans can package complicated plots into a relatively small set of well-recognized and meaningful linguistic terms. This summarization ability though has not been available to systems that deal with narrative and would be important in creating higher quality systems. In this paper, we describe preliminary work towards a machine learning model of plot summarization using conditional random fields and describe our own feature functions inspired by cognitive theories of narrative reasoning. Our approach allows us to learn summarization models of single character event driven narratives and automatically summarize new narratives later on.

This paper examines davidreaming: spontaneously recalling or imagining personal or vicarious experiences in the East or future. The following functions of daydreaming, for oth humans and computers, are discussed: support for processes of creativity, future planning and rehearsal, learning from successes and failures, emotion modification, and motivation. A computational theory of daydreaming is being implemented and tested in a computer program called DAYDREAMER. A prototype version of DAYDREAMER which produces several daydreams {in Engiish) is currently running.