Ingrid Zukerman

AIJ Journal 2011 Journal Article

The automated understanding of simple bar charts

• Stephanie Elzer
• Sandra Carberry
• Ingrid Zukerman

While identifying the intention of an utterance has played a major role in natural language understanding, this work is the first to extend intention recognition to the domain of information graphics. A tenet of this work is the belief that information graphics are a form of language. This is supported by the observation that the overwhelming majority of information graphics from popular media sources appear to have some underlying goal or intended message. As Clark noted, language is more than just words. It is any “signal” (or lack of signal when one is expected), where a signal is a deliberate action that is intended to convey a message (Clark, 1996 [15]). As a form of language, information graphics contain communicative signals that can be used in a computational system to identify the message that the graphic conveys. We identify the communicative signals that appear in simple bar charts, and present an implemented Bayesian network methodology for reasoning about these signals and hypothesizing a bar chart's intended message. Once the message conveyed by an information graphic has been inferred, it can then be used to facilitate access to this information resource for a variety of users, including 1) users of digital libraries, 2) visually impaired users, and 3) users of devices where graphics are impractical or inaccessible.

IJCAI Conference 2009 Conference Paper

• Fabian Bohnert
• Daniel F. Schmidt
• Ingrid Zukerman

Spatial processes are typically used to analyse and predict geographic data. This paper adapts such models to predicting a user’s interests (i. e. , implicit item ratings) within a recommender system in the museum domain. We present the theoretical framework for a model based on Gaussian spatial processes, and discuss efficient algorithms for parameter estimation. Our model was evaluated with a real-world dataset collected by tracking visitors in a museum, attaining a higher predictive accuracy than state-of-the-art collaborative filters.

IJCAI Conference 2007 Conference Paper

• Yuval Marom
• Ingrid Zukerman

We are developing a corpus-based approach for the prediction of help-desk responses from features in customers' emails, where responses are represented at two levels of granularity: document and sentence. We present an automatic and human-based evaluation of our system's responses. The automatic evaluation involves textual comparisons between generated responses and responses composed by help-desk operators. Our results show that both levels of granularity produce good responses, addressing inquiries of different kinds. The human-based evaluation measures response informativeness, and confirms our conclusion that both levels of granularity produce useful responses.

IJCAI Conference 2005 Conference Paper

A Probabilistic Framework for Recognizing Intention in Information Graphics

• Stephanie Elzer
• Sandra Carberry
• Ingrid Zukerman
• Daniel Chester
• Nancy Green
• Seniz

This paper extends language understanding and plan inference to information graphics. We identify the kinds of communicative signals that appear in information graphics, describe how we utilize them in a Bayesian network that hypothesizes the graphic’s intended message, and discuss the performance of our implemented system. This work is part of a larger project aimed at making information graphics accessible to individuals with sight impairments.

AAMAS Conference 2005 Conference Paper

Voting Policies that Cope with Unreliable Agents

• Christian Guttmann
• Ingrid Zukerman

AAMAS Conference 2003 Conference Paper

An Information-theoretic Approach for Argument Interpretation in a Conversational Setting

• Sarah George
• Ingrid Zukerman
• Mark George

IJCAI Conference 1999 Conference Paper

Exploratory Interaction with a Bayesian Argumentation System

• Ingrid Zukerman
• Richard McConachy
• Kevin Korb
• Deborah Pickett

We describe an interactive system which supports the exploration of arguments generated from Bayesian networks. In particular, we consider key features which support interactive behaviour: (1) an attentional mechanism which updates the activation of concepts as the interaction progresses; (2) a set of exploratory responses; and (3) a set of probabilistic patterns and an Argument Grammar which support the generation of natural language arguments from Bayesian networks. A preliminary evaluation assesses the effect of our exploratory responses on users' beliefs.

IJCAI Conference 1999 Conference Paper

Pre-sending Documents on the WWW: A Comparative Study

• David Albrecht
• Ingrid Zukerman
• Ann Nicholson

Users' waiting time for information on the WWW may be reduced by pre-sending documents they are likely to request, albeit at a possible expense of additional transmission costs. In this paper, we describe a prediction model which anticipates the documents a user is likely to request next, and present a decisiontheoretic approach for pre-sending documents based on the predictions made by this model. We introduce two evaluation methods which measure the immediate and the eventual benefit of pre-sending a document. We use these evaluation methods to compare the performance of our decision-theoretic policy to that of a naive pre-sending policy, and to identify the domain parameter configurations for which each of these policies provides a clear overall benefit to the user.

AAAI Conference 1998 Conference Paper

Bayesian Reasoning in an Abductive Mechanism for Argument Generation and Analysis

• Ingrid Zukerman

Ourargumentation system, NAG, uses Bayesian networks in ausermodelandin a normative modelto assemble andassess arguments whichbalance persuasiveness withnormative correctness. Attentional focusis simulated inbothmodels to selectrelevantsubnetworks for Bayesian propagation. The subnetworks areexpanded in aniterativeabductive process until argumentative goalsareachieved in bothmodels, when theargument ispresented to theuser.

AIJ Journal 1998 Journal Article

Inductive learning of search control rules for planning

• Christopher Leckie
• Ingrid Zukerman

One method for reducing the time required for plan generation is to learn search control rules from experience. The most common approach to learning search control knowledge has been explanationbased learning. An alternative approach is to use inductive learning. An inductive approach does not require a complete and tractable domain theory and has the potential to create more effective rules by learning from more than one example at a time. In this paper we describe Grasshopper, an inductive system that learns search control rules for a classical plan generation system. We also provide an empirical evaluation of Grasshopper by comparing it with an existing explanation-based learning system.

UAI Conference 1997 Conference Paper

Lexical Access for Speech Understanding using Minimum Message Length Encoding

• Ian E. Thomas
• Ingrid Zukerman
• Jonathan J. Oliver
• David W. Albrecht
• Bhavani Raskutti

The Lexical Access Problem consists of determining the intended sequence of words corresponding to an input sequence of phonemes (basic speech sounds) that come from a low-level phoneme recognizer. In this paper we present an information-theoretic approach based on the Minimum Message Length Criterion for solving the Lexical Access Problem. We model sentences using phoneme realizations seen in training, and word and part-of-speech information obtained from text corpora. We show results on multiple-speaker, continuous, read speech and discuss a heuristic using equivalence classes of similar sounding words which speeds up the recognition process without significant deterioration in recognition accuracy.

ECAI Conference 1996 Conference Paper

Constraint Propagation in a Cooperative Approach for Multimodal Presentation Planning

• Yi Han 0001
• Ingrid Zukerman

ECAI Conference 1996 Conference Paper

Using Argument Graphs to Generate Arguments

• Richard McConachy
• Ingrid Zukerman

ECAI Conference 1994 Conference Paper

Acquisition of Information to Determine a User's Plan

• Bhavani Raskutti
• Ingrid Zukerman

ECAI Conference 1994 Conference Paper

Being Concise versus Being Shallow: Two Competing Discourse Planning Paradigms

• Ingrid Zukerman
• Richard McConachy

AAAI Conference 1993 Conference Paper

An Optimizing Method for Structuring Inferentially Linked Discourse

• Ingrid Zukerman

In recent times, there has been an increase in the number of Natural Language Generation systems that take into consideration a user’ s inferences. The statements generated by these systems are typically connected by inferential links, which are opportunistic in nature. In this paper, we describe a discourse structuring mechanism which organizes inferentially linked statements as well as statements connected by certain prescriptive links. Our mechanism first extracts relations and constraints from the output of a discourse planner. It then uses this information to build a directed graph whose nodes are rhetorical devices, and whose links are the relations between these devices. The mechanism then applies a search procedure to optimize the traversal through the graph. This process generates an ordered set of linear discourse sequences, where the elements of each sequence are maximally connected. Our mechanism has been implemented as the discourse organization component of a system called WISHFUL which generates concept explanations.

UAI Conference 1991 Conference Paper

Handling Uncertainty During Plan Recognition in Task-Oriented Consultation Systems

• Bhavani Raskutti
• Ingrid Zukerman

AAAI Conference 1987 Conference Paper

Goal-Based Generation of Motivational Expressions in a Learning Environment

• Ingrid Zukerman

In a tutorial setting, we often hear expressions such as "The method we are about to discuss will help you solve . . . " or "Let us consider a subject which demands some more practice," which are issued by a tutor to motivate a student to attend to forthcoming discourse. In this paper we model the meaning of these expressions in terms of their anticipated influence on the status of a listener’s goals, and use these predictions to produce motivational expressions and embed them in computer generated discourse. In particular, we have recognized relations which are instrumental in determining a listener’s motivational requirements in a hierarchical problem-solving domain. These ideas have been incorporated into a system called FIGMENT which generates commentaries on the solution of algebraic equations.

AAAI Conference 1986 Conference Paper

Comprehension-Driven Generation of Meta-Technical Utterances in Math Tutoring

• Ingrid Zukerman

A technical discussion often contains conversational expressions like "however," "as I have stated before," "next," etc. These expressions, denoted Meta-technical Utterances (MTUs) carry important information which the listener uses to speed up the comprehension process. In this research we model the meaning of MTUs in terms of their anticipated effect on the listener comprehension, and use these predictions to select MTUs and weave them into a computer generated discourse. This paradigm was implemented in a system called FIGMENT, which generates commentaries on the solution of algebraic equations.