HCI Bibliography Home | HCI Conferences | IUI Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
IUI Tables of Contents: 9397989900010203040506

Proceedings of the 1993 International Workshop on Intelligent User Interfaces

Fullname:International Workshop on Intelligent User Interfaces
Editors:Wayne D. Gray; William E. Hefley; Dianne Murray
Location:Orlando, Florida
Dates:1993-Jan-04 to 1993-Jan-07
Standard No:ACM ISBN 0-89791-556-9; 0-89791-557-7 (hard); ACM Order Number 608930; ACM DL: Table of Contents hcibib: IUI93
  1. Plenary Sessions
  2. Session 1: Intelligent Presentation
  3. Session 2: Modeling
  4. Session 3: Intent Recognition
  5. Session 4: Adaptive Systems
  6. Session 5: Dialog Management/Natural Language
  7. Session 6: User Support
  8. Session 7: Design & Evaluation
  9. Short Papers
Intelligent User Interfaces BIBAKPDF 3-10
  William E. Hefley; Dianne Murray
We consider the significant work in both the human-computer interaction and artificial intelligence fields in the area of intelligent interfaces, placing into a structure the disparate strands of research work. We treat design and implementation issues, covering the relationship between user interface management systems (UIMS) architectures and interface design, and providing schematic architectures for building intelligent systems. We discuss such work in its historical context and suggest that UIMS architectures be used as a basis for constructing intelligent interfaces.
Keywords: Intelligent interfaces, Intelligent user interfaces, User models, User interface management systems, Adaptive intelligent interface systems

Plenary Sessions

The Future of Intelligent Interfaces? Not Just "How?", But "What?" and "Why?" BIBAPDF 13-17
  Ernest Edmonds
The paper argues that the future development of intelligent interfaces depends as much upon understanding the scope and purpose of their intelligent behaviour as upon developing the mechanisms that operate them. It reviews the different ways in which an interface can be intelligent, what it can be intelligent about, for what purpose and under whose control. The dimensions of intelligence include data to human action, command syntax to organizational context and reduce error to become rich.
   We know the solution, but what's the problem?
The Price of Flexibility BIBAPDF 19-25
  David D. Woods
Flexibility and customizability are central to the perceived advantages of the growth in technological powers. But this and other vectors of technological change, when used clumsily, create new burdens and complexities for beleaguered human practitioners responsible for achieving goals within some field of activity. "Intelligent interfaces" are sometimes seen as a solution to the growing demands of highly technological and highly automated fields of activity. However, data from a variety of sources indicates that a technology driven approach to the development of intelligent interfaces is likely to provide the illusion of assistance while creating a new layer of burdens and complexities. What distinguishes technological flexibilities that are used to increase the range of practitioner adaptive response to the variability resident in the field of activity from technological flexibilities that simply create new burdens on the practitioner especially at high tempo, high criticality periods?

Session 1: Intelligent Presentation

Knowledge-Based Design of Graphical Responses BIBAKPDF 29-36
  Helen Chappel; Michael Wilson
Articulate graphics systems choose the most effective form of presentation of data rather than being tools forcing users to design graphics themselves. An architecture for a graphic response designer is proposed which clearly separates the graphical design knowledge from knowledge of the user, current task, application and dialogue context which are required to tailor responses. Without this division the graphic response designer cannot be a portable module and will duplicate resources available elsewhere in an intelligent interface. Graphical design rules derived from design practice are described as well as the method for integrating these with the supporting knowledge. An example intelligent interface incorporating a graphic response designer in conjunction with natural language output is described. The problems of updating the graphic design knowledge and evaluating the demonstrator are considered.
Keywords: Articulate graphics, Intelligent interfaces, User modelling, Automatic graphic presentation
Supporting Interactivity in Automated 3D Illustrations BIBAKPDF 37-44
  Doree Duncan Seligmann; Steven Feiner
An interactive intent-based illustration is a picture designed to satisfy an input communicative intent, and which can be interactively redesigned as it is viewed. We describe how the architecture of IBIS (Intent-Based Illustration System) automates the design of 3D interactive intent-based illustrations. The types of interaction that IBIS supports include changes in the world, as objects move and otherwise change state; changes in the communicative intent, as modified by the user or other programs; queries, in which the user can request additional information; and self-evaluation, in which other programs can request IBIS to analyze various properties of the illustration, such as how well an illustration's communicative intent is satisfied or which graphical techniques it uses. We show how IBIS's design process and rule base make possible these forms of interaction.
Keywords: Intelligent user interfaces, Generation of nonverbal media, Automatic presentation, Explanation, Knowledge-based graphics
Some Aspects of the User Interface of a Knowledge Based Beautifier for Drawings BIBAKPDF 45-52
  Dieter Bolz
Automatic beautification of drawings is a technically interesting problem. Equally attractive is the problem of how such a service can be presented at the user interface of a graphics editor, and, conversely, how to restrict the functionality to what is presentable. We have implemented a (semi-) automatic beautifier for drawings as an extension to a conventional graphics editor. In this report, we present a close look at the current implementation of the beautifier's user interface and give an overview of further research activities.
Keywords: Knowledge based systems, Critique, Beautifying

Session 2: Modeling

Intelligent Mediation: An Architecture for the Real-Time Allocation of Interface Resources BIBAKPDF 55-61
  Russell Ovans; William S. Havens
Operator interfaces to supervisory-control systems are often highly complex, cumbersome to extract information from, and overwhelmingly verbose in the face of abnormal operating conditions. An oft-cited solution is to replace the conventional operator interface with an intelligent interface; one that mediates control system output to present the operator with intelligently formatted information. While an appealing idea, the question of how to do this remains unanswered: intelligent presentation of control system data is a difficult problem. The task is complex because it requires the real-time allocation of limited interface resources. An expert system architecture and methodology -- called intelligent mediation -- for the real-time allocation of limited interface resources is proposed as a solution to this problem.
Keywords: Supervisory control, Constraint logic programming, Knowledge-based interface design, Automatic presentation architectures
Knowledgeable Development Environments Using Shared Design Models BIBAKPDF 63-70
  Robert Neches; Jim Foley; Pedro Szekely; Piyawadee Sukaviriya; Ping Luo; Srdjan Kovacevic; Scott Hudson
We describe MASTERMIND, a step toward our vision of a knowledge-based design-time and run-time environment in which human-computer interfaces development is centered around an all-encompassing design model. The MASTERMIND approach is intended to provide integration and continuity across the entire life cycle of the user interface. In addition, it facilitates higher quality work within each phase of the life cycle. MASTERMIND is an open framework, in which the design knowledge base allows multiple tools to come into play and makes knowledge created by each tool accessible to the others.
Keywords: Models, Collaboration, Design, Development
The Study of Models of Intelligent Interfaces BIBAKPDF 71-78
  Angel R. Puerta
Researchers in the field of intelligent interfaces have concentrated on building architectures, and have placed little emphasis on defining appropriate models. As a result, this research area is not well defined. L-CID is a model of an intelligent interface that establishes the knowledge requirements for, determines the functionality of, and creates a definition for an intelligent interface. In addition to modeling knowledge-based user interaction, L-CID allows two important functions commonly overlooked in models of intelligent interfaces: (1) self-adaptation, and (2) user-interface management. Examples of implementations of L-CID for each of these functions are presented.
Keywords: Intelligent-interface modeling, Self-adaptation, User-interface management, Machine learning

Session 3: Intent Recognition

A Learning Interface Agent for Scheduling Meetings BIBAKPDF 81-88
  Robyn Kozierok; Pattie Maes
This paper describes a Learning Interface Agent for a meeting scheduling application. The agent employs Machine Learning techniques to customize itself to the user's personal scheduling rules and preferences by observing the user's actions and receiving direct user-feedback. Our approach provides the user with sophisticated control over the gradual delegation of scheduling tasks to the agent, as a trust relationship is built. We report upon an experiment in which a collection of such assistants became gradually more helpful to their users through the use of memory-based and reinforcement learning. The experimental data reported upon demonstrate that the learning approach to building intelligent interface agents is a very promising one which has several advantages over more standard approaches.
Keywords: Interface agents, Learning interface agents, Machine learning, Personal assistants, Software agents
Plan Recognition for Context Sensitive Help BIBAKPDF 89-96
  Klaus-Jurgen Quast
In this paper we describe a technique for modeling action plans by multilayered symbolic nets. The nets are used as a signal conductive system by a newly developed inference component in order to recognize the modeled action plans. Using spreading activation releases the knowledge base designer from handling interrupts which can disturb a recognition process. Since the inference process can be stopped, continued and backed up without completely reseting and recomputing the inference status, this technique can be used to model UNDOs. The inference component generates an action history which stores recognized executions of actions and subactions on each level. This history can be evaluated by other components. We describe an application for an EXCEL help system to offer the user a selection of help topics sensitive to his or her task handling. The Evaluator analyses the computed history in order to determine crucial usage problems and detect suboptimal task executions which occur using an application program.
Keywords: Modeling action plans, Recognition of action plans, Spreading activation inference, History of plan executions, Context sensitive help
The Role of Built-in Knowledge in Adaptive Interface Systems BIBAKPDF 97-104
  Daniel Crow; Barbara Smith
We discuss the construction of task-oriented interfaces and argue that they must adapt themselves to each individual user's behaviour. Because of the variation between users, it is impracticable to attempt to build in a priori assumptions about the user. We present an adaptive interface system, DB_Habits, which shows that it is possible, however, to incorporate knowledge about the underlying system, and which uses this to recognise repeated sequences of commands issued by the user which represent the user's higher-level tasks. DB_Habits uses the command sequences found, to collaborate with the user in achieving the user's tasks and is malleable i.e. can be easily adapted by the user to support their tasks. We present results showing the effect of incorporating simple command syntax knowledge on the performance of the system.
Keywords: Adaptive user interfaces, Task-oriented interfaces, User variation, Pattern recognition, User modelling

Session 4: Adaptive Systems

Supporting Adaptive Interfaces in a Knowledge-Based User Interface Environment BIBAKPDF 107-113
  Piyawadee "Noi" Sukaviriya; James D. Foley
Developing an adaptive interface requires a user interface that can be adapted, a user model, and an adaptation strategy. Research on adaptive interfaces in the past suffered from a lack of supporting tools which allow an interface to be easily created and modified. Also, adding adaptivity to a user interface so far has not been supported by any user interface systems or environments.
   In this paper, we present an overview of a knowledge base model of the User Interface Design Environment (UIDE). UIDE uses the knowledge of an application to support the run-time execution of the application's interface and provides various kinds of automatic help. We present how the knowledge model can be used as a basic construct of a user model. Finally, we present adaptive interface and adaptive help behaviors that can be extended to the current UIDE architecture utilizing the user model. These behaviors are options from which an application designer can choose for an application interface.
Keywords: User interface design environment, User model, Adaptive interface, Adaptive help
Developing Adaptive Systems to Fit Individual Aptitudes BIBAKPDF 115-121
  David Benyon; Dianne Murray
We focus on designing systems which can adapt to individual differences in personality and cognitive style. To do so, we need to identify individual cognitive and personalily characteristics, validate them and discover appropriate design solutions to deal with such differences. These factors are long-term and relatively stable individual aptitudes and therefore are difficult for humans to change. We believe that it is exactly these sort of differences with which adaptive systems should concern themselves. We describe how we have approached identification of these characteristics and describe how they were incorporated in an operational, though functionally quite simple system.
Keywords: Adaptive interfaces, User models, Individual differences
An Adaptive Environment for the User Interface of Excel BIBAKPDF 123-130
  Christoph G. Thomas; Mette Krogsoeter
Software systems developed in recent years are becoming increasingly powerful, but in most cases they tend to abandon the user to deal with the complexity of the system alone. There is an immense need for systems with individual, context-sensitive support. In this paper, we discuss the development of such a support system with special focus on the user interface. We describe the extension of an existing and complex program, Excel, to an adaptive and customizable system. This extension is called Flexcel and has been implemented and evaluated.
   The primary aim of this extension is to provide a convenient environment for adapting Excel's user interface to particular users and their current tasks. For example, it allows the user to define new menu entries for actions that are frequently needed and normally require a tedious dialog step. Flexcel also analyzes the user's interaction style and presents adaptation suggestions. A critique component supports the user in his problem solving and learning activities related to the adaptability of Flexcel's user interface.
Keywords: Adaptivity, Customizable, Critique, Knowledge bases, Usage profile

Session 5: Dialog Management/Natural Language

Development of the INRS ATIS System BIBAKPDF 133-140
  Changwen Yang; Douglas O'Shaughnessy
We are establishing a dialogue-by-voice system which is a spoken language interface to the Official Airline Guide (OAG) Database for the general public to book air tickets directly with computer databases (including over the telephone). It consists of an ATIS (Air Travel Information System) natural language system, an automatic SQL (Standard Query Language) generator, a speech recognizer, a speech synthesizer, a text generator, an N-best algorithm and an N-best interface.
   This paper describes our successfully developed modules: (1) ATIS natural language system which can process not only an isolated imperative, question, assertion, but also a complete booking scenario; (2) an automatic SQL generator which has been connected to the ATIS natural language system to realize generic public transactions with a computer database through natural language sentences; (3) an N-best interface which guarantees that the ATIS natural language system will accept N sentence candidates sent from the speech recognizer.
Keywords: Natural language processing, Spoken language system, Syntax, Semantics, Pragmatics
Commenting on Action: Continuous Linguistic Feedback Generation BIBAKPDF 141-148
  Wim Claassen; Edwin Bos; Carla Huls; Koenraad De Smedt
Action mode interfaces, in which users achieve their goals by manipulating representations, suffer from some fundamental disadvantages. In this paper, we present a working prototype of a system called Continuous Linguistic Feedback Generator (CLFG), a facility that addresses the major disadvantages. CLFG generates natural language descriptions of the actions the user is performing. These descriptions are presented in both the visual and audio channels. The knowledge sources and algorithm that enable CLFG to provide relevant and concise information are described in detail.
Keywords: Natural language generation, Multimodal interfaces, Action mode
Knowledge Bases for User Guidance in Information Seeking Dialogues BIBAKPDF 149-156
  Anne Tissen
Information seeking activities are inherently interactive processes. Modeling the interaction gives the chance to build flexible interfaces, which are adaptable to the user's tasks and goals, but at the risk of loosing orientation in the information space. We propose to guide the information seeking process by a structured dialogue control. Applying a case-based approach, we employ cases of previous successful dialogues, which are adaptable to the new problem situation, for guiding and controlling an information seeking dialogue. It has become evident that flexible dialogue behavior of interface systems can be achieved only if background knowledge is modelled in the system explicitly. We present several knowledge bases about dialogue structure, the user's tasks and goals, and the system's knowledge about the domain and how they fit together during the ongoing process. Our approach is implemented in a prototype of a case-based dialogue manager (CADI), applied in the MERIT interface system to the CORDIS databases.
Keywords: Modeling techniques for interface design, User guidance, Case-based dialogue manager, Information retrieval

Session 6: User Support

Explanation Facilities and Interactive Systems BIBAKPDF 159-166
  Hilary Johnson; Peter Johnson
Our main research aim is to improve the provision of explanation facilities in information systems generally, and to identify what is meant by "explanation". This paper reports research which identifies both the strengths and weaknesses of current research and shows how to overcome those weaknesses. We are also concerned with both present and future uses of explanation in information systems and the role of explanation in a broad range of interactive applications.
Keywords: Explanation, Explanation theories, Task-based explanation, Task knowledge structures (TKS)
Providing Opportunistic Enrichment in Customized On-Line Assistance BIBAKPDF 167-174
  Ursula Wolz
Regardless of the sophistication of an intelligent interface, a user must develop expertise in using it. This requires that the interface include the ability to both respond informatively to users questions and opportunistically enrich user expertise. If the interface is not able to take the initiative in providing enrichment, then the user must decide when and how to learn more about the system. This paper presents a solution to the problem of how to provide opportunistic enrichment. An interface called GENIE that answers user queries through natural language generation techniques is presented. In answering users' questions, GENIE is able to provide enrichment in the following forms. It can introduce new material. It can clarify distinctions between a method (plan) for doing a task that is the focus of the query and one that is not. It can also elucidate misconceptions that are not the focus of the query.
Keywords: Automated assistance, Intelligent computer assisted instruction, User modeling, Natural language generation
Generating Explanations in Context BIBAKPDF 175-182
  Giuseppe Carenini; Johanna D. Moore
If user interfaces are to reap the benefits of natural language interaction, they must be endowed with the properties that make human natural language interaction so effective. Human-human explanation is an inherently incremental and interactive process. New information must be highlighted and related to what has already been presented. In this paper, we describe the explanation component of a medical information-giving system. We describe the architectural features that enable this component to generate subsequent explanations that take into account the context created by its prior utterances.
Keywords: Explanation, Dialogue management, Tutoring systems, Natural language processing

Session 7: Design & Evaluation

Cognitive Task Analysis, Interface Design, and Technical Troubleshooting BIBAKPDF 185-191
  Linda S. Steinberg; Drew H. Gitomer
We propose a model of interface design that makes use of two interdependent levels of cognitive analysis: 1) the study of the criterion task through an analysis of expert/novice differences and; 2) the application of a GOMS analysis to a working interface design. We review this dual analysis in the context of HYDRIVE, a video-disc based intelligent tutoring system designed to facilitate the development of troubleshooting skills for aircraft hydraulics systems. The initial cognitive task analysis enabled the identification of critical troubleshooting skills and troubleshooting procedures. We find, though, that even with an in-depth initial cognitive task analysis, the GOMS interface analysis resulted in significant and beneficial design changes.
Keywords: Cognitive task analysis, Hydraulics maintenance, Intelligent tutoring system, Interface design, Troubleshooting
Wizard of Oz Studies -- Why and How BIBAKPDF 193-200
  Nils Dahlback; Arne Jonsson; Lars Ahrenberg
We discuss current approaches to the development of natural-language dialogue systems, and claim that they do not sufficiently consider the unique qualities of man-machine interaction as distinct from general human discourse. We conclude that empirical studies of this unique communication situation is required for the development of user-friendly interactive systems. One way of achieving this is through the use of so-called Wizard of Oz studies. We describe our work in this area. The focus is on the practical execution of the studies and the methodological conclusions that we have drawn on the basis of our experience. While the focus is on natural language interfaces, the methods used and the conclusions drawn from the results obtained are of relevance also to other kinds of intelligent interfaces.
Keywords: Design and evaluation, Dialogue, Natural language interfaces
Towards Automatic Evaluation of Multimodal User Interfaces BIBAKPDF 201-208
  Sandrine Balbo; Joelle Coutaz; Daniel Salber
The evaluation of the usability and the learnability of a computer system may be performed with predictive models during the design phase. It may be done on the executable code as well by observing the user in action. In this case, data collected in vivo must be processed. Our goal is to provide a software support for performing this difficult task.
   This article presents an early analysis and experience towards the automatic evaluation of multimodal user interfaces. With this end in view, a generic Wizard of Oz platform has been designed to allow the observation and the automatic recording of subjects' behavior while interacting with a multimodal interface. We then show how recorded data can be analyzed to detect behavioral patterns, and how deviations of such patterns from a data flow-oriented task model can be exploited by a software usability critic.
Keywords: Capture of behavioral data, Multimodal user interface, Wizard of Oz, User interface evaluation techniques

Short Papers

A Prototype Intelligent User Interface for Real-Time Supervisory Control Systems BIBAKPDF 211-214
  Iain M. Begg; Joe Gnocato; Wendy E. Moore
This poster describes a prototype intelligent user interface for real-time control systems. The prototype combines advanced computer graphics visualization technology with expert systems and human factors engineering to produce an Intelligent Graphic Interface for operators of real-time supervisory control systems, such as those used in telecommunications network management, power distribution and industrial process control. We describe the project, domains, architecture and implementation of the first proof-of-concept prototypes.
Keywords: Intelligent user interface, Intelligent mediation, Real-time control, Network management, Power systems
Reducing User Effort in Collaboration Support BIBAKPDF 215-218
  Andy Cockburn; Harold Thimbleby
The value of electronic mail as a medium for collaborative and coordinated work can be enhanced by relating messages to conversations. While some groupware systems have offered such facilities, their ability to assess conversational context is dependent on explicit user action and the use of specific systems by all collaborators.
   This paper describes Mona, a novel conversation based email platform. Mona provides a hypertext representation of conversational context without requiring any additional effort from the user or the use of specific email systems by other collaborators. Mona's lack of requirements and independence is made possible by inferring conversational context with heuristics using information inherently transferred in all email communications.
   Mona's heuristics are described, as are its mechanisms for personalising conversation views.
Keywords: Email, Free guidance, Conversational context, Heuristics, Mona (a CSCW system)
The Automation of Helpdesks BIBAKPDF 219-222
  L. M. Coventry; T. B. Kane
Helpdesks are becoming increasingly more important in large organisations. These organisations may be commercial or academic. However, management of such helpdesks is becoming increasingly more difficult as the amount of knowledge that must be acquired and maintained by the advisor becomes rapidly unmanageable. The sources of information are highly distributed and the domains are becoming larger. No one person can be an expert in all necessary domains. Many attempts are being made to increase the effectiveness of these helpdesks. The solution may lie in an intelligent assistant for the advisor. This paper describes such a system. The advisors remain central to the process. They may train the system and their ability to communicate effectively with a variety of users, each with different experience and knowledge, modes of learning and information requirements cannot be replicated by a software system. However, an intelligent system can help the advisor to maintain their knowledge, acquiring knowledge from a variety of sources and assisting in the diagnosis of problems. The advisor formulates queries from users for the system, receives the information from the system and communicates this in a manner appropriate to the user who had the initial problem.
Keywords: Agents, Blackboard architecture, Advisory system, Interface, Answer filters
Experimental Results on User Knowledge Assessment with an Evidential Reasoning Methodology BIBAPDF 223-226
  Michel C. Desmarais; Jiming Liu
User modeling is a crucial element of intelligent interfaces. This paper presents some preliminary results on a method for the assessment of user knowledge. Knowledge of an individual is modeled as a subset of a global set of knowledge units (KU). The KUs are interrelated among each other through implication relations, indicating that certain KUs are prerequisite to others. The structure of implications is induced algorithmically through an induction procedure utilizing empirical data on the knowledge state of a number of subjects. It is thereafter used in a knowledge assessment process based on the Dempster-Shafer evidential reasoning methodology. The preliminary results demonstrate that our approach can successfully assess knowledge by randomly sampling KUs in a knowledge state and infer which KUs are known and unknown.
Adaptive Function Allocation by Intelligent Interfaces BIBAKPDF 227-229
  P. A. Hancock; M. H. Chignell
Intelligent interfaces act as translational intermediaries between humans and machines while themselves representing a third component in the human-interface-machine triad. They seek to maximize the collaborative capabilities of human and machine through the communication of mutual queries and actions in symbols best suited to the respective entity involved. This triad is directed to the achievement of a range of external goals. Attainment of such goals is best served by the enhancement of mutual adaptive capability of the individual components of the human-interface-machine triad and of the triad itself as a supraordinate unit. Within collective action, the question of who does what, when and with whom is best seen as a problem in adaptive allocation. This question is itself a subset of the general problem of function allocation which is arguably the foundation of human-machine interaction. This work advocates modular customization of the intelligent interface as a solution to the impasse that derives from previous and failed descriptive approaches to the allocation question.
Keywords: Task allocation, Intelligent interface
Inferring Complex Plans BIBAKPDF 231-234
  Kristina Höök; Jussi Karlgren; Annika Wærn
We examine the need for plan inference in intelligent help mechanisms. We argue that previous approaches have drawbacks that need to be overcome to make plan inference useful. Firstly, plans have to be inferred -- not extracted from the users' help requests. Secondly, the plans inferred must be more than a single goal or solitary user command.
Keywords: Plan inference, Interactive help systems, Configuration problem
An Ecological Man-Machine Interface for Temporal Visualization BIBAKPDF 235-238
  Lars Peter Jensen; Peter Koch
This paper concerns the design of man-machine systems, especially the interface between man and machine (MMI).
   Three different models of human information processing are used to extract theoretical guidelines for designing MMI's to support human mental activity. The models, all developed by Professor Jens Rasmussen Denmark, are: Skill-, Rule- and Knowledgebased behaviour, Abstraction hierarchies and Decision ladder.
   Based on the extracted guidelines for ecological man machine interface a new viewport for both overview and detailed information is developed: The Spider display.
   The display is presented and explained and a link to the 3 models is made showing that the Spider display can be used as a tool for the system designer attempting to design an ecological display, which match the operator's way of thinking and acting.
Keywords: Interface-building tools and techniques, User modelling techniques for ecological interface design
Cooperative Support for Distributed Supervisory Control: Requirements, Issues, and an Example from Mission Operations BIBAKPDF 239-242
  Patricia M. Jones
The term distributed supervisory control emphasizes that supervisory control is commonly conducted by teams of human operators. This paper focuses on issues in distributed supervisory control, particularly with regard to providing cooperative problem solving tools, and provides an example real-world application.
Keywords: Supervisory control, Activity management, Computer-supported cooperative work
A User-Centered Approach to Adaptive Interfaces BIBAKPDF 243-245
  Thomas Kuhme
An adaptive user interface is supposed to adapt itself to the characteristics of an individual user. It is widely accepted that such an adaptation requires the interface to maintain a user model embedded in the system. In this article an inspectable user model is proposed which gives the user an insight into adaptation strategies and underlying assumptions. It supports the user in adapting the interface and serves as an instrument for controlling automatic adaptations. The inspectable user model plays a central role in a user-centered approach to adaptive user interfaces.
   This approach, called Computer-Aided Adaptation (CAA), focuses on enabling users to express their needs and preferences more effectively. It aims to provide a better transparency of the adaptation process and underlying user modeling mechanisms. The user model is expected to provide the means for an implicit adaptation. And finally, the adaptation process itself should be made adaptive due to the users' individual prerequisites concerning adaptation.
   The ideas presented in this paper are the basis of a CAA environment being under development.
Keywords: Adaptive user interfaces, User modeling
Adjusting the Presentation of Forms to Users' Behavior BIBAKPDF 247-249
  Uwe Malinowski
This poster presents an approach to make form-based interfaces easier to use. The presentation of forms is adapted to the individual user's behavior. Color and highlighting of items are adjusted with respect to the time of last change of the item and the usually used item value compared to the currently used value. By this means complex forms can be used effectively, as the user will be able to identify relevant information at first glance. It is discussed what information about users' behavior has to be represented in a user model in order to provide for adaptation.
Keywords: Intelligent user interfaces, User modelling, Adaptive systems
Issues in Practical Application of an Adaptive Interface BIBAKPDF 251-254
  Beth Meyer; K. C. Burgess Yakemovic; Michael Harris
The authors have been developing a prototype system for installation in an operational business environment. During the development, a number of issues have been encountered. These include:
  • constraints arising from placing prototypes in operational environments
  • lack of guidelines for selecting types of adaptation
  • difficulty determining adaptation criteria
  • difficulty obtaining necessary data from users
  • lack of guidelines for making information display decisions
  • problems in testing 'real world' interfaces These issues are not readily solved with more sophisticated adaptation algorithms; rather, they point out the need for collecting more information from attempts to bring intelligent interfaces to complex business environments.
    Keywords: Adaptive interfaces, Prototypes, Commercial applications, Development issues
  • Focus in Graphical User Interfaces BIBAKPDF 255-257
      Manuel A. Perez; John L. Sibert
    Focus in natural language processing is used to keep track of the attentional space of the participants in the dialogue. Graphical interfaces have still to benefit from the use of this level of communication. We discuss a graphical interface counter part called a focus space. Focus spaces can be used to combine multiple interaction styles (e.g. natural language with graphical pointing). We describe how a specific theory of discourse structure and focus space can be applied to graphical interfaces. The attentional space is maintained automatically by the system, and can be used by the software under program control. An example is provided showing how the attentional space can be used.
    Keywords: Focus, Discourse structure, Graphical user interfaces
    Creating Intelligent User Interfaces Using Prototyping and Knowledge Based Support Technologies -- The Rapid Prototyping Tool MacEMSIG -- BIBAKPDF 259-262
      Stephan Romahn; Annette Kaster
    The rapid prototyping tool MacEMSIG supports the designer of user interfaces with extensive interactive editors, so the graphical layout creation process is simple and easy. Since the interface layout is assembled completely separated from the application system, the interface can be iteratively modified as well as linked to either system simulations or real systems. MacEMSIG has a runtime environment that allows realtime operations with the application system where user activity can be recorded. The data may be used for empirical interface analysis or for on-line user support. MacEMSIG aids in the creation of various operation concepts, such as integrated, system-oriented and task-oriented designs. Knowledge-based systems can be used to assist the user to perform the task. The simulation of the Airbus Flight Management System and an user interface to a real system are presented as examples.
    Keywords: Human-machine systems, User interface design, Artificial intelligence, Knowledge-engineering, Simulation
    A User Interface for the Electronic Encyclopedia Exploratorium BIBAKPDF 263-267
      Michael Salisbury; Alan Borning
    The Electronic Encyclopedia Exploratorium (E{cubed}) is a vision of a future computer system -- a kind of electronic "How Things Work" book. Typical articles in E{cubed} will describe such mechanisms as compression refrigerators, engines, telescopes, and mechanical linkages. Each article will provide simulations, three-dimensional animated graphics that the user can manipulate, laboratory areas that allow a user to modify the device or experiment with related artifacts, and a facility for asking questions and receiving customized, computer-generated English-language explanations. During the past year we have implemented an initial prototype of an article on refrigerators and are currently implementing a more sophisticated second version. In this short paper we focus on the user interfaces for these prototypes.
    Keywords: "How things work" articles, Simulations, Animation, Laboratory areas, Refrigerators
    Programming with Characters BIBAPDF 269-272
      Mike Travers; Marc Davis
    Programs are hard to build, and even harder to understand after they are built. We lack intuitive interfaces for visualizing and manipulating many parts of programs and the ways in which these parts interact. Constraint systems have addressed these problems. We generalize some of the notions inherent in constraint systems to agent-based systems, and explore the use of animated characters as interface representations of agents. In particular, conflict detection and resolution is dramatized by the use of characters and their emotions. The history of their interactions is presented as a narrative using video and storyboard techniques. Building programs out of agents and enabling users to manipulate program parts by interacting with simple animated characters can aid relatively unskilled users in understanding and modifying complex systems.