HCI Bibliography Home | HCI Conferences | EWHCI Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
EWHCI Tables of Contents: 919293

East-West International Conference on Human-Computer Interaction: Proceedings of the EWHCI'93

Fullname:East-West International Conference on Human-Computer Interaction
Editors:Len Bass; Juri Gornostaev; Claus Unger
Location:Moscow, Russia
Dates:1993-Aug-03 to 1993-Aug-07
Publisher:International Centre for Scientific and Technical Information (ICSTI) 125252, 21-B Kuusinen Str. Moscow, RUSSIA
Standard No:hcibib: EWHCI93
  1. EWHCI 1993-08-03 Volume 1
    1. Teaching and Learning
    2. Hypertext
    3. Foundations of HCI
    4. Empirical Studies
  2. EWHCI 1993-08-03 Volume 2
    1. Information Visualization/Navigation
    2. Interface Design: Techniques, Tools and Paradigms
    3. Customizing Interfaces
    4. Multi-Media
  3. EWHCI 1993-08-03 Volume 3
    1. Empirical Studies
    2. Experience/Applications
    3. Expert Systems/Knowledge Engineering

EWHCI 1993-08-03 Volume 1

Teaching and Learning

HyperLecture: A Self-Organizing Lecture Presentation and Revision System BIBA 11-26
  Damian Conway
This paper presents an overview of HyperLecture, a hypertextual, gesturally-controlled lecture presentation system originally designed for teaching introductory computer programming. HyperLecture provides genuinely user-friendly mechanisms with which a presenter can quickly produce linear and non-linear presentations, hard-copy hand-outs, audio and textual annotations, complete self-driving interactive tutorials and graphical programming code simulations.
Towards an Adaptive Hypermedia Component for an Intelligent Learning Environment BIBA 27-38
  Peter Brusilovsky; Leonid Pesin; Mikhail Zyryanov
This paper discusses the problem of integration of hypermedia and Intelligent Learning Environments (ILEs) technologies and the problem of creating an adaptive hypermedia component for ILEs. Our experience of creating an adaptive on-line help facilities for ITEM/IP system is described. This experience forms a background for our hypermedia work and provides some good ideas for it. We also present our approach to integration of a hypermedia component into internal structure of ILE and illustrate it with two examples of adaptive hypermedia components for the most recent versions of our ILEs ISIS-Tutor and ITEM/PG. Finally, we summarize main features of this approach, provide some references to related works, and consider some issues of adaptive hypermedia in general.
Teaching Art History through Multimedia Systems: Museum's Experience BIB 38-44
  Olga Kissel
An Intelligent Interface for Computer Assisted Language Learning BIBA 45-58
  Eve Wilson
This paper begins by contrasting teacher directed computer assisted language learning with the student directed approaches of recent hypertext learning programmes. It concludes that what is needed is an approach where the level and order of the exercises can be tailored to the needs and requirements of individual students. To do this requires:
  • 1. a means of constructing an original user profile and of updating this in the
        light of student performance in the tutorial exercises, and
  • 2. a means of generating exercises of an appropriate level from text databases. The paper next looks at how texts may be automatically assigned a readability grade based on Gunning's Fog Index or Information Density, and how this can be used to generate exercises of varying levels of difficulty. The paper emphasizes the importance of giving students feedback to win their co-operation in devising and adhering to a tuition programme. It concludes by discussing the work still to be done.
  • Optimum Teaching System -- Illusion or Reality? BIB 59-78
      Yuriy Souldin


    The Challenge of Effectively Integrating Graphics into Hypertext BIBA 78-86
      Keith Instone; Erik Brown; Laura Leventhal; Barbee Teasley
    It is important that designers of hypertext do not assume that the way information is presented in traditional media such as paper books will necessarily be effective in hypertext. Several studies have shown that graphics presented with text on a screen are not necessarily used effectively, or even used at all. In our study, we explored three ways of presenting textual and graphical information about geographical locations and objects. In all cases, the user had to point with a mouse to a item name in order to see text, and the text appeared near the mouse cursor. In one presentation style, no graphic was present. In a second style, a graphic was present, but the text appeared as a caption. In the hypertext style, the user pointed to item names on the graphic. The results showed that the hypertext style resulted in about 20% better learning of both spatial information (which was illustrated by the graphic) and of information not illustrated by the graphic.
    A Hypertext Model with Partly Defined Nodes BIBA 87-92
      Nicholas Sidorov; Svetlana Gorobets; Vladimir Kopach
    From the predicate approach is presented a model of hypertext that admits of creation and treating of nodes which are using the partly defined information objects. The described model of hypertext is realized and good used in the technology of the software recovery for the solution of the problem of understanding the source code for example for autobuilding of hypertext representance of the source programmes' texts.
    INTELTEXT: Producing Coherent Linear Texts While Navigating in Large Non-Hierarchical Hypertexts BIBA 93-102
      Martin Subbotin; Dmitry Subbotin
    Inteltext is a new software technology able to construct automatically a unified text from separate text items. The main advantage of Inteltext is its ability to support those whose main work is thinking and exposing thoughts in textual form. Inteltext is capable to draw out of large heap of information and represent in logical form text items which can complement or confirm your thought. The paper describes the underlying approach being developed in Russia since 70-s and the software product implementing it.
    Coherent Navigation in Hypertext Environments: Using a Computer-Shared Macrocontrol Based on Structural Strategies of Discourse Comprehension and Production (The SMIsC Conception) BIBA 103-122
      Valery M. Chelnokov; Victoria L. Zephyrova
    The key usability problem in hypertext and hypermedia is the Navigation Problem being subdivided into the problem of disorientation and the problem of cognitive overhead. We believe that one systematic approach to the NP consists in developing mechanisms assisting users to navigate coherently, i.e. to blaze trails imitating semantically coherent discourses in hypertext/hypermedia networks. The main purpose of a mechanism of such kind is maintenance the macrocontrol over local transitions among nodes of information during a navigational (interactive) session. This macrocontrol should provide trail's global (thematic) coherence, that is its meaning integrity.
       Basing theoretical part of our work upon the works by such known psycholinguists as van Dijk, Kintsch and Levelt, we demonstrate that a good portion of macrocontrol maintenance may be placed onto computer -- namely, the portion we describe as navigational strategies which are content-independent, being formulated in terms of network's graph node-link structure only, and are expressible in the form of machine instructions. The mechanisms of strategies we have developed aids users to move in the network, maintaining trail's global coherence in relation to a chosen theme and imitating a so called canonically ordered discourse from the network's nodes. With this mechanism, the macrocontrol becomes shared among computer and user and this makes the user's burden being significantly eased. We use the term 'System of Meaning Integrities structural Creation', or SMIsC, to denote a hypertext system with such shared macrocontrol.
    HyperMethod: An Intelligent Interface for Hypertext BIBAK 123-131
      Igor V. Kotelnikov; Dmitry L. Krechman
    The basis for human (mode of) thinking is formed by several foreformal mechanisms which manifest themselves in the form of algorithms and specifications in any sphere of human activity. The purpose of the paper is to suggest realization models for some of mechanisms and to show that hypertext may be a special non-program shell or environment inside which a user can solve a wide range of specific problems.
    Keywords: Hypertext, Graph, Object oriented programming, Logic programming, Relation algebra

    Foundations of HCI

    Historical Analysis and Conflicting Perspectives -- Contextualizing HCI BIBA 132-142
      Susanne Bødker
    This paper develops two ways of analyzing the human-computer interaction of a computer application in use in an organization. The techniques, historical analyses and conflicting perspectives analysis, and the interplay between them, are used in providing the basis for a more detailed analysis.
        Historical analyses focus on the historical development of artifacts and their use. Conflicting perspectives analysis reflects on the roles of the artifact in use, as system, tool, or medium. Combined, the two types of analysis allow for a focus in particular on conflicts between the roles of a specific artifact in use.
        The techniques are based on human activity theory. They are illustrated by means of a case study of a computer application from a project with the Danish National Labour Inspection Service.
    A Brief Look at the Psychological and Linguistic Foundations of HCI BIBA 143-159
      Boris Enikeev; Yuriy Kaganov; Dmitriy Zhuk
    In this paper the HCI is regarded as humanities problem. At this time IT and HCI is one of the most important component of the contemporary civilization. Its importance determines orientation to human activities. Because IT and HCI requires humanities analysis. We are going to use some of the concepts cognitive psychology, semiotics, structural linguistics for analysis HCI.
    Visual User Interfaces: A Psychophysiologist's Perspective BIBA 159-163
      Vladimir M. Krol
    In report describe theoretical thesis and experimental results demonstrating necessity using in advanced user's interface of the knowledge about input characteristics of the visual system of the man. The new generation of the visual user's interface should be based also on knowledges about work of the inherent mechanisms visual thinking.
    Hierarchical Components of Human-Computer Systems: Issues BIBA 163-179
      Mark Sh. Levin
    This paper describes the hierarchical components of human-computer system (HCS). The development and utilization of three packages for IBM PC, i.e., DSS 'COMBI' for multicriteria ranking, hierarchical hypertext system 'HHS' with main orientation for multicriteria evaluation and the DSS for hierarchical design 'SED' are under consideration. The study is based on an analysis of HCS components (information, user, techniques) and major operations (development, representation, correction, learning and using).
    Iconic Signs and Languages in User Interface Design BIBA 180-190
      Vladimir Andreev; Gennady Uzilevsky
    Various aspects of complex problem of icon construction for user interface design are considered, ergosemiotic requirements to icons design are revealed, features and functions of iconic languages are defined.
    Impediments to the Process of Group Work BIBAK 191-207
      Beth Adelson
    In this paper we begin by presenting a taxonomy of impasses in group work situations. The taxonomy includes factors such as goal conflicts and resource limitations. We then present a prescriptive theoretical framework designed to support negotiation during these impasses. We also describe NegotiationLens, a system which embodies the framework by supporting the actions prescribed by the theory. We then analyze the adequacy of the framework which stresses a collaborative form of negotiation. From this analysis we suggest a line of research which would lead to an expanded taxonomy. We hypothesize the expanded taxonomy would include interpersonal factors such as inequalities in the power of negotiating parties. We then discuss the framework and tools which would be useful given this expanded view of causes of group work impasses. Lastly we suggest the relevance of these factors to other classes of groupware.
    Keywords: Group work/CSCW, Interaction architectures, Structure of organizations, Negotiation support tools/environments
    Computer-Aided Support of Cooperative Group Design BIB 207-218
      I. V. Kotenko; O. E. Ryabov
    Interacting with Proprietary Software Users: An Application for Activity Theory? BIBA 219-226
      Insiah Bourke; Irina Verenikina; Edward Gould
    A much neglected but vitally important section of software users are demanding to be noticed. These can be loosely defined as proprietary users and range in skill from computer novices to experts. This paper looks at the characteristics of this group and asks if Activity Theory offers them, and those who manage them, any hope for a better deal.

    Empirical Studies

    Query Composition: Why Does It Have to be So Hard? BIBA 226-241
      Lucy Terry Nowell; Deborah Hix; Eric D. Labow
    Project Envision, a large research effort at Virginia Tech, focuses on developing a user centered, multimedia database from the computer science literature, with full-text searching and full-content retrieval capabilities. User interviews indicate that people have trouble composing queries. Widely available boolean retrieval systems present problems with both syntax and logic. Natural language queries for vector space retrieval systems are easier to compose, but users complain that they do not understand the matching principles used; users also complain that they have too little control over the search and fear being overwhelmed by an enormous retrieval set. We describe the Envision query window, which has as a usability goal making query composition easy while increasing user control. Results of formative usability evaluation and subsequent redesign are discussed.
    Interface Semantics and Procedural Knowledge: A Study of Novice Understanding of MacDraw BIBA 241-256
      M. V. Springett; A. S. Grant; A. G. Sutcliffe
    This paper presents an analysis of issues affecting the choice of usability evaluation strategies for highly interactive Direct Manipulation interfaces. It reports memory tests on novice users of MacDraw I which investigate the degree of reliance on the interface for cueing, and the sources of information, both within the interface and outside, that may be utilized. A study of the novice subjects' (verbalised) reasoning during experimental task-performance on MacDraw is then reported. Examples of subjects reasoning are used to illustrate the nature of Direct Manipulation evaluation. Issues affecting the selection and development of evaluation methods are then discussed.

    EWHCI 1993-08-03 Volume 2

    Information Visualization/Navigation

    Language Expressiveness in Software Visualization Systems BIBA 5-8
      V. L. Averbukh; I. V. Tarskikh
    The conceptions of the visual expressiveness and the user interface expressiveness are suggested. Problems of Software Visualization system evaluating are discussed.
    E{cubed}: Towards the Metrication of Graphical Presentation Techniques for Large Data Sets BIBA 9-26
      Ying K. Leung; Mark D. Apperley
    Rapid advances in communications and computer technologies in recent years have provided users with greater access to large volumes of data from computer-based information systems. Whilst researchers have developed many novel techniques to overcome the problems associated with the presentation and navigation of large data sets on a limited display surface, the choice of a technique in a particular application remains very subjective. This paper proposes an evaluation framework E{cubed} which aims to provide a basis for the comparison of different presentation techniques, given the nature and characteristics of the data to be presented, and the interpretation required. E{cubed} focuses on three aspects of graphical data presentation: expressiveness, efficiency, and effectiveness. This framework lays the foundation for the development of a set of metrics to facilitate an objective assessment of presentation techniques.
    Navigating in a Process Landscape BIBA 26-38
      Haakon Tolsby
    User interfaces in process control systems are often organized as landscapes -- usually a process chart -- where the display is a window into the landscape. The window can move over the landscape and view different segments of the process. Normally there is no overlap between the views -- they are discretely organized. Operators often claim they get lost in such interfaces. I suggest that the problem is a matter of how the operator moves in the landscape -- how one changes ones view of the process. Two different interactive methods of movement -- discrete vs. continuous -- are tested. The results of the experiments are discussed.
    Visualisation of Complex Information BIBA 38-50
      Matthew Chalmers
    In information retrieval, sets of documents are stored and categorised in order to allow for search and retrieval. The complexity of the basic information is high, with representations involving thousands of dimensions. Traditional interaction techniques for such complex information therefore hide much of its complexity and structure, and offer access to it by means of isolated queries and word searches. Bead is a system which takes a complementary approach, as it builds and displays an approximate model of the document corpus in the form of a map or landscape constructed from the patterns of similarity and dissimilarity of the documents making up the corpus. In this paper, emphasis is given to the influences on and principles behind the design of the landscape model and the abandonment of a 'point cloud' model used in an earlier version of the system, rather than the more mathematical aspects of model construction.
    Three-Dimensional Visualisation of Knowledge Structures: Prototyping for Design Evaluation BIBA 50-70
      U. K. Patel; A. G. Sutcliffe
    A conceptual framework for specification of 3D visualisations is introduced, and a task knowledge modelling approach to designing 3D visualisations of knowledge structures is described. Using this approach, a three-dimensional interactive graphical user interface to a medical diagnosis knowledge based system has been designed and specified. The design has been implemented as a prototype and evaluated. We found that there are individual differences in the way users explore three dimensional visualisations, and that usability is dependent on both the morphology (visualisation) and manipulations (interface functionality). Implications of these findings for the design of 3D visualisations are discussed.
    Visual Programming in R-Technology: Concepts, Systems and Perspectives BIBA 71-88
      Igor Ushakov; Igor Velbitskiy
    This paper represents the efficient framework for software development environments based on a R-charts and gives main concepts and aims of a project directed to the development of the visual user interface for C++ programming language. R-charts are structured graphs which are used for visual specifications of algorithms, visual programming and debugging. They have been applied to different languages and have adopted by ISO as an alternative charting notation for the program constructs in ISO/IEC 8631. The user interface for C++ described in this paper incorporates the R-charts, OO diagrams and other visualization methods for visual specifications, data browsing and control flow visualization.

    Interface Design: Techniques, Tools and Paradigms

    Coupling Interaction Specification with Functionality Description BIBA 88-97
      A. Kameas; S. Papadimitriou; G. Pavlides
    In this paper, the solution used in the context of SEPDS (a Software Development Environment) to the problem of combining interactive behavior specification with functionality description of a distributed interactive application is presented. This solution consists of combining two specification models: IDFG to describe the interactive aspects of applications developed with the system and EDFG to describe their functionality. Both these models are data flow graph based and can be classified as process models. They use "actors" to represent performers of processes and "links" to represent data buffering and exchange, as well as roles and different perspectives. Although the two models have many semantical differences, they also have many common properties, that is why they can be straightforwardly combined in a process that enables designers think in users terms. To this end, action actors are used to represent the functions supported by the application, and context actors to represent the application user interface functions. In addition, links are used to represent the events that take place in the system (these may be user or system actions), the effects that these have on the screen, the context into which these take place and the goals that may be achieved using the application. Furthermore, the reusability and prototyping tools of SEPDS can be used to construct and test the application design.
    WYSIWYG Editors: And What Now? BIBAK 98-115
      Eddy Boeve; Lon Barfield; Steven Pemberton
    Most editors nowadays are said to be WYSIWYG ('What you see is what you get'). Although this implies that the effects of user actions are made immediately visible to the user, this does not usually include the effects of other causes. This a logical consequence of the fact that the user edits a copy of the document, rather than the document itself. These kind of systems then, can better be classified as 'What you see is what you will get' systems.
       This report describes an editor model that is a further extension of the WYSIWYG principle: 'Things are exactly as they appear', or TAXATA for short. In these kind of systems, the user carries out every action by editing, and what is more important, by editing the object directly. Furthermore, modifications made to objects by the system are made immediately visible to the user.
       Amongst other things, the reports describes the underlying model and the necessary editing concepts to construct such a TAXATA editor environment, based on general user-interface principles. Finally the design of one particular edit command has been described, to give an impression of the specific design issues in such an environment.
    Keywords: Software engineering, Tools and techniques, User interfaces, Software engineering, Program coding, Program editors, Software engineering, Programming environments, Interactive, User-interfaces, Syntax directed editors, Editor design
    An Extension to the Human-Computer Interaction Paradigm BIBA 115-123
      R. C. MacGregor
    This paper suggests that the focus of user support has moved from a simple system-user interface to a need to consider a variety of potential interfaces at the work-setting. It argues that the current HCI paradigm still only addresses user support in terms of a technological interface, but that this must be extended to include a number of other alternatives.
    Designing Multimedia Interfaces BIBA 123-133
      Alistair Sutcliffe; Peter Faraday
    Multimedia interfaces are currently created by intuition. Development of a method for analysis and design of multimedia presentation interfaces is described. The study investigates task based information analysis, persistence of information, selection attention and concurrency in presentation. The method gives an agenda of issues, diagrams and techniques for specification, and guidelines for media selection and presentation scripting. Use of the method is illustrated with a case study of shipboard emergency management.
    Synthesis-Oriented Situational Analysis in User Interface Design BIBAK 134-150
      Kevin A. Mayo; H. Rex Hartson
    Analytic evaluation is a term describing a class of techniques for examining a representation of a user interface design, and discovering design flaws and/or predicting user task performance. In our work with analytic evaluation, we have observed limitations on the effectiveness and efficiency of analytic techniques for formative evaluation supporting the iterative design and re-design cycle. Here we support those observations with arguments based on theoretical limitations of the models underlying these techniques. By way of comparison we discuss desirable characteristics for an alternative approach. In our search for such an alternative, we have developed the Task Mapping Model, a substantively different approach to analysis for supporting the user interface design. We briefly describe the Task Mapping Model and give some examples illustrating its desirable characteristics.
    Keywords: User interface design, User interface design requirements, User interface evaluation, Task description, Task analysis
    Architecture Elements for Highly-Interactive Business-Oriented Applications BIBA 151-173
      Francois Bodart; Anne-Marie Hennebert; Jean-Marie Leheureux; Benoit Sacre; Isabelle Provot; Jean Vanderdonckt
    It is now widely recognized that powerful architecture elements are needed for implementing the whole lifecycle (specification and design) od interactive applications. In this paper, we deal with the application model of the TRIDENT project which introduces three components: the application component, the dialog component and the presentation component. This hierarchical object-oriented architecture rests on the use of four kinds of objects: application objects, dialog objects (whose one particular includes the supervisory object) and interaction objects. An abstract data model and selection rules for choosing appropriate interaction objects according to the application's semantic and to the user level are given.
    Three Interface Levels BIB 174-176
      M. V. Donskoy

    Customizing Interfaces

    Personal Counselor: A Hybrid Adaptive User Interface BIBA 177-193
      Max M. North; Sarah M. North
    This paper discusses a hybrid adaptive interface, called Personal Counselor, for tolerating an individual user's error behavior. The fusion of different adaptive interface techniques, such as adaptive scheduling, self-adaptive, and teaming agent, can provide a powerful hybrid adaptive interface for many domain-specific problems. Specifically, Personal Counselor adapts itself to individuals by observing their actions and receiving direct feedback. At the beginning, Personal Counselor requires minimal background knowledge, but gradually this knowledge becomes more sophisticated and more reliable. The authors present an application of Personal Counselor to Unix operating system commands. The preliminary results of the pilot studies suggest subjects who utilize Personal Counselor perform at higher level (t=6.85, df=78, p<0.05) than their counterparts who did not utilize the interface.
    Generating Self-Adaptive Human-Computer Interfaces BIBA 193-202
      Aline Montoy-Berthome
    Self-adaptive interfaces can be considered at different levels: adaptation to the user's habits and experience, adaptation to the user's task, adaptation to the user's knowledge. In this paper, we explain what we mean by adapting to the user's experience and task, how we detect the need of such adaptations and how we realize them, and why we have been led to consider automatic generation of interfaces. We expose then what we aim to do about adaptation to the user's knowledge.
    The Practical Use of Macro Recording: A Case Study BIBAK 203-210
      Allen Cypher
    Macro recording is a practical technique for automating repetitive tasks on computers. The user records a series of actions, and then the computer can re-execute those actions. This paper discusses a variety of macros that were used to assist in a real-life task of editing a book manuscript. The capabilities of current tools are presented, followed by a discussion of how the limitations of these tools restrict current end users, and how some of these limitations can be overcome.
    Keywords: Macro recording, Automation, End user programming, Demonstrational interfaces, Programming by demonstration
    Addressing Usability by Customising Interfaces for Organisations BIBAK 210-220
      Colin Tattersall; Peter Plantfeber
    New telecommunications services are becoming available which will provide a high level of functionality to a wide range of customers. This degree of utility will have to be coupled with ease of use in order to capture a share of the increasingly competitive telecommunications market. Usability will become a "buy/don't buy" factor for customers faced with services from different telecommunications companies which offer similar levels of functionality.
       This paper argues that usability issues can only be addressed with respect to the particular contexts-of-use to which a service will be put, and that services should be offered with interfaces which are designed to reflect this use within organisations. That is, a generic service should be offered with an interface specialised to the needs of particular vertical markets or organisations.
       An approach to the design of services is presented which exploits separability of interface and application to allow tailored services to be delivered to customers.
    Keywords: Usability, Interfaces, Organisations


    The MSM Framework: A Design Space for Multi-Sensori-Motor Systems BIBA 220-232
      Joelle Coutaz; Laurence Nigay; Daniel Salber
    One of the new design goals in Human Computer Interaction is to extend the sensory-motor capabilities of computer systems to better match the natural communication means of human beings. This article proposes a dimension space that should help reasoning about current and future Multi-Sensori-Motor systems (MSM). To do so, we adopt a system centered perspective although we draw upon the "Interacting Cognitive Subsystems" psychological model. Our problem space is comprised of 6 dimensions. The first two dimensions deal with the notion of communication channel: the number and direction of the channels that a particular MSM system supports. The other four dimensions are used to characterize the degree of built-in cognitive sophistication of the system: levels of abstraction, context, fusion/fission, and granularity of concurrency. We illustrate the discussion with examples of multimedia and multimodal systems, both MSM systems but with distinct degrees of built-in cognitive sophistication.
    Radiological Reporting Based on Voice Recognition BIBA 233-246
      G. Antoniol; R. Fiutem; R. Flor; G. Lazzari
    Speech recognition has proved to be a natural interaction modality and an effective technology for medical reporting, in particular in the radiology speciality. High-volume text creation needs and the complex structure of these texts make voice technologies useful. By employing speech, professionals in the field can generate reports and do so at a speed that approaches traditional dictation methods.
       However, the integration of speech recognition in a user interface creates new problems: speech recognizers may introduce errors and moreover they should be adaptable to spoken language variations.
       This paper describes a radiological reporting system and the related motivations for the use of the speech modality. A preliminary evaluation of the system has shown that, on average, although text recalling functions and keyword shortcuts are available, more than two thirds of a radiological report are generated by means of dictation.
    Task Interference with a Discrete Word Recognizer BIBA 246-252
      Caryn Hubbard; James H. Bradford
    Speaker dependent, discrete word recognition is the simplest and most successful form of automatic speech recognition. In the near future, it is likely that this technique will be the basis for a variety of commercial speech interfaces. However, discrete word recognition requires users to insert relatively long pauses between each word of an utterance. This paper describes an experiment that was performed to determine whether this unusual way of speaking will interfere with the performance of complex tasks.
    Model of Utterance and Its Use in Cooperative Response Generation BIBA 252-264
      Koichi Yamada; Riichiro Mizoguchi; Naoki Harada; Akira Nukuzuma; Keiichi Ishimaru; Hiroshi Furukawa
    A cooperative response model is proposed for interactive intelligent systems that recognizes user intentions and makes cooperative responses. Though many models developed so far have shown that they can achieve some form of cooperative responses, the coverage of each model is limited. In this paper, we propose a model which covers various types of cooperative responses. The paper starts with a classification of cooperative responses and discusses the relation between intentions and responses. Based on the discussion, a user utterance model is introduced and an intention recognition mechanism is developed employing domain-independent rules and knowledge about the normal usage of the topic object. The recognized intentions are then used to generate appropriate cooperative responses.
    Special Computer Interfaces for the Visually Handicapped: F.O.B. the Manufacturer BIBA 265-273
      Arthur I. Karshmer; Richard L. Oliver
    Many techniques have been suggested, and some even brought to market, to allow the visually handicapped person to more easily interact with modern computing equipment. Most of the work to date has focused on providing special purpose hardware and software to accomplish this task. In the current work, we describe an approach that would allow all computer manufacturers to ship systems based on today's popular graphical user interfaces (GUIs) that will also serve the needs of the visually handicapped user. By building the user interface into the GUI normally supplied by the manufacturer, the cost of such interfaces should go down, while the availability should go up.

    EWHCI 1993-08-03 Volume 3

    Empirical Studies

    Experts' Assessments of Robustness, Stability, and Sources of Expertise in the Programming Process BIBA 5-17
      Brad Blumenthal
    Expert programmers have a great deal of knowledge about their code that is not directly related to its structure or function, and they use this knowledge extensively while programming. While modern programming environments capture a great deal of formal knowledge about a piece of code, they do not capture the informal knowledge that programmers use while coding. The problem with building environments to capture such informal knowledge is determining what knowledge to capture. Although there are a variety of studies on programmers's representations of their code, there are few, if any, conclusions about what informal representations expert programmers use when working with large software systems. This paper presents conclusions from a study indicating that expert programmers have consistent assessments of such characteristics as robustness (the reliability of a piece of code), stability (the amount that a piece of code has changed recently), and sources of expertise (the people who know about a piece of code and can explain it). In addition, programmers use these assessments of robustness, stability, and sources of expertise to guide the debugging, design, and modification processes. These results have significant implications for the design of new programming environments, and based on these implications, some new directions for studying expert programmers and building programming environments are discussed.
    Structuring the Field of HCI: An Empirical Study of Expert Representations BIBA 18-28
      Peter Brusilovsky; Ivan Burmistrov; Victor Kaptelinin
    In this paper we present results of empirical study of HCI field based on the free sorting technique with subsequent cluster analysis. Eight proven HCI experts participated in the study proposing classifications of papers presented at the EWHCI'92 conference. The results showed satisfactory degree of consensus between the experts and high level of interpretability of group data. Some important findings are discussed. The results of our study could be useful for HCI conference organizers in their activity on arrangement of papers to special conference sessions.
    Testing ADAT -- An Automated Debugger for Ada Tasks BIBA 29-44
      Arthur V. Lopes; Thomas J. Nagy
    This paper describes the evaluation of ADAT (Automated Debugger for Ada Tasks). As a concurrent Small Ada program compiles, it generates a higher level abstraction of the program's intertask interactions. These abstractions are entered into CLIPS in the form of facts. Next, the CLIPS inference engine matches the facts against the rules which emulate the debugging knowledge of an expert Ada programmer. For cases where mismatches exist, ADAT issues an explanation of the error and its location and, when possible, ADAT also provides guidance on how to correct the problem. The details of the above individual steps are hidden from the user. ADAT was developed using a combination of Small Ada and CLIPS.
    Positive Test Bias in Software Testing among Professionals: A Review BIBA 45-54
      Laura Marie Leventhal; Barbee M. Teasley; Diane S. Rohlman; Keith Instone
    Fundamental but virtually unexplored issues in human-computer interaction involve the roles of biases in software engineering tasks. In studies of naturalistic testing tasks, as well as ones which follow common laboratory models in this area, we have found ample evidence that testers have positive test bias. This bias is manifest as a tendency to execute about four times as many positive tests, designed to show that "the program works," as tests which challenge the program. In our prior work, we have found that the expertise of the subjects, the completeness of the software specifications, and the presence/absence of program errors may reduce positive test bias. Skilled computer scientists invent specifications to test in the absence of actual specifications, but still exhibit positive test bias.
    Applying the Wizard of Oz Technique to the Study of Multimodal Systems BIBA 55-67
      Daniel Salber; Joelle Coutaz
    The Wizard of Oz (WOz) technique is an experimental evaluation mechanism. It allows the observation of a user operating an apparently fully functioning system whose missing services are supplemented by a hidden wizard. From our analysis of existing WOz systems, we observe that this technique has primarily been used to study natural language interfaces. With recent advances in interactive media, multimodal user interfaces are becoming popular but our current understanding on how to design such systems is still primitive. In the absence of generalizable theories and models, the WOz technique is an appropriate approach to the identification of sound design solutions. We show how the WOz technique can be extended to the analysis of multimodal interfaces and we formulate a set of requirements for a generic multimodal WOz platform. The Neimo system is presented as an illustration of our early experience in the development of such platforms.
    A Sociological and Psychological Portrait of a Computer Network User BIBA 68-75
      Olga Arestova; Leonid Babanin; Alexander Voiskounsky
    The article concerns the results of the research on Relcom (the largest computer network accessible in Russia) users. The parameters chosen for research include demographical, sociological and psychological ones. A portrait of a typical user is based on the results of research.


    A System to Model, Assist and Control the Human Observation of Microscopic Specimen BIBA 75-85
      A. Derder; C. Garbay
    This paper describes current research on computerized assistance to cytological specimen exploration. The purpose is not to design a new diagnosis expert system, but rather to design a system able to cooperate with the human expert in the execution of specimen exploration task. New man machine assistance models are necessary to this end, which imply not only knowledge-based but also behaviour-based modeling. An information manager is described, allowing access to information supplied by this model. An error monitoring is also presented. Its goal is to control all tasks and activities involved in the cytological specimen exploration.
    On Applications of the Statistical Software System "CLAVDIA" to Some Problems of Medical Diagnostics BIBA 85-100
      Vladimir S. Stepanov; Eduard A. Gukasian; Boris N. Filatov; Svetlana P. Grabovskaja
    The interactive system CLAVDIA (Cluster-Analysis, Visualization, Discriminant Analysis) belongs to the group of the method-oriented intellectualizing statistical software developed for the IBM PC platform.
       The system is intended for classification, reduction of dimensionality and some accompanying problems. In this paper, the system has been applied for statistical processing of the data containing some indices of people's health for the following problems:
  • - differential diagnostics of a lungs empyema form;
  • - diagnostics of the type of people's poisoning with some medical drug;
  • - analysis of health indices of people which have worked beside Chernobyl
       Nuclear Power Plant.
  • Task-Oriented Robot Control BIBA 100-104
      A. S. Jushenko; I. V. Maslov
    Task-oriented robot control is performed by task-oriented program intelligent interface between robot operator and robot control subsystem. This paper provides you with main aspects of development of task-oriented interface.
    Software Advertising in the Press BIBA 105-108
      Alexander E. Voiskounsky; Anna V. Kuznetsova
    The paper deals with a qualitative analysis of software advertisements published in Russian newspapers. The major trends, the benefits and disadvantages of software advertizing are taken into consideration.
    Problems in Elaborating Computing Technology in the Agricultural Business BIB 109-112
      B. S. Kasaev
    The "Document-Driven Activity" Approach to Modelling and Designing Office Automation Systems BIBA 112-120
      Alexander Chernin
    A "document-driven activity" approach to office automation (OA) system design is proposed. According to the approach, each action with documents is activated by changes in documents states. The use of Petri nets for modelling and prototyping of OA systems is also suggested. The approach allows a system analyst to incorporate the main features of the office technology and to ensure the correctness of OA system structure and algorithms.
    Development of an Instrument for Human Organism Complex Diagnostic BIB 120-123
      Igor Scorodumov

    Expert Systems/Knowledge Engineering

    Computer-Aided Knowledge Engineering: An Interface-Centred Approach BIBA 123-129
      Tatjana Gavrilova; Andrey Zolotarev; Sergey Yeremenko
    The paper briefly traces the history of the knowledge acquisition tools development and describes features and reasons of new approach to this problem -- computer aided knowledge engineering (CAKE). First a brief analysis of the most popular tools in this field is presented.
    DI*GEN: The Shell for Building of Diagnostic Expert Systems BIBA 129-132
      Serge Greenberg; Tatyana Yakhno
    To build an expert system is not an exotic art anymore. Now it is obviously a well-organized activity for many specialists that have little in common with programming. To provide them with good instruments, a wide spectrum of the technological methods and expert systems building tools (ESBT) are developed including so-called shells.
       This paper describes the shell DI*GEN that is under development in the Artificial Intelligence Laboratory of the Informatics Systems Institute in Novosibirsk and is intended for building of diagnostic expert systems.
    The Efficient Implementation of Declarative Knowledge Representation in a Problem-Oriented Expert System Shell BIBAK 132-136
      Serge Greenberg; Tatyana Yakhno
    In the paper the structure of the DI*GEN shell is described. This shell is intended for building diagnostic expert systems as everyday well organized activity.
       The main idea of the DI*GEN is that at the knowledge acquisition stage all problem knowledge is described by expert in the object-oriented style with help of special tools. Then the rule-based representation which is used on the consultation stage is generated automatically and is hidden from expert. We suggested an approach of complete compilation of knowledge base into procedural code as the alternative to interpretive approach for reasoning.
    Keywords: Diagnostic systems, Object-oriented knowledge acquisitions, Compilation of the knowledge base
    An Expert System for Tutoring in Information Networks BIB 136-138
      Andrew Dziengelewski; Victor Rumyancev; Aleksew Zarovny
    An Interactive Computer Simulation of Catastrophic Earthquake Cycles BIBA 138-140
      Grigoriy Kazakevich; Leopold Lobkovsky; Vladimir Tochitsky
    To forecast earthquakes it is necessary to know the mechanism of their development. The seismotectonic process in island arcs and in active continental margins can be described with the help of the "keyboard" model of L. I. Lobkovsky [1] of great earthquakes, that is based on modern geodynamics conceptions. An interactive computer software is designed based on this model. It facilitates calculations for the regions, the user is interested in, and refinement of the model parameters. The presented interactive software is intended for use in research, applications and educational programs. The "keyboard" model can also be extended to intracontinental earthquakes by using two-level plate tectonics concepts (see [1]).