Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 56

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53

Strict Standards: Only variables should be assigned by reference in /home/projects/cs-www/wwwcs/groups/IK/index.php on line 53
Center for Content and Knowledge Engineering (CCKE)

Printer Friendly Version
Webmaster
J. van den Berg
   Home         |    Raison d'Etre         |    Positioning         |    Focus         |    Projects         |    People         |    Faq         |    Contact          

 

  Chair Electronic Document Technology
  Chair Cognition and Communication
 

Projects in the area of Cognitive and Communicative Aspects of Information Technology and Information Systems


Within the cognition aspect we work on the following projects:

  • Externalization of Information (Tabachneck-Schijf, H. van Oostendorp, C. van Nimwegen (Ph.D.student)).A recurrent issue for interaction design is how much information should be made available on the interface (externalization of information) as opposed to having users encode and store this information in memory (internalization). We investigate in this project conditions whereby externalizing the information in the interface (e.g., by using visualization) or requiring it to be internalized, respectively, enhances the task performance of users. Results support that requiring internalization is beneficial to the user, rather than externalizing all needed information. In particular, the long-term mental model of the problem at hand improves, facilitating transfer and solving the problem again after some months. This research contributes to formulate guidelines for designing graphical user interfaces in the context of problem solving tasks
  • Cognitive Model for Navigating on the Web (I. Juvina (Ph.D. student), M.C. Puerta, H. van Oostendorp). The aim of this project is to build a cognitive model that can predict and explain human navigation performance in Web-assisted tasks. In the first phase of research, user, interface and context characteristics that are relevant to task outcomes have been empirically identified. On this basis we are developing a cognitive model of web navigation. The current version of the model accounts for aspects of user web navigation behavior such as: backtracking actions, considering contextual information when judging goal relevance, and employing navigation strategies. Two empirical studies have shown that when users are provided with model-generated suggestions of navigating actions, the number of navigating steps and backtracking actions decreases whereas task performance improves. In the next phase, the model is refined and implemented as a computational cognitive model, and the value of the model in generating adaptive navigation support is examined.We also examine the relevance of navigation metrices such as 'lostness' as predictor of task performance.
  • Navigation Support on Mobile Devices (S. Nagata (Ph.D. student), in cooperation with TNO HF). In collaboration with TNO HF we are developing a framework for realizing assistants that offer support in navigating Web-based services on mobile devices. The focus of our project is on how to handle and mediate interruptions while performing web-based tasks in a mobile context. Two empirical experiments have shown that device interruptions (i.e. Instant Messaging) versus interruptions from the environment (e.g. cell phones, paging) are highly disruptive to a user on a mobile device influencing web task performance. The repetitiveness of an interruption task (e.g. short bursts of communication during an Instant Messaging session) can also negatively impact performance. Whereas the anticipation of an interruption and maintaining the same modality when switching between an interruption and web task is less damaging to user handling of interruptions during a mobile web task. We are developing general support principles for handling the interruptions and empirically testing their value.

  • Engineering the Affective Appraisal of 3D Models of Buildings(J. Houtkamp, Ph.D project, H. van Oostendorp, in collaboration with Prof. M. Overmars). The process of modeling a 3D building and displaying the model on a monitor cause modifications in the representation of the building that distort affective appraisal. These must be compensated by manipulations of the model to restore such appraisals. In this AGS project - funded by GATE- we try to develop a systematic approach to engineering the requirement of affective appraisal, which comprises analyzing the building and the 3D model of that building, manipulating elements of the model, and testing the results. For instance, can we, by modeling the design of a railway station, assess how safe and comfortable people will feel there? This project extends the cognitive effects of interaction to the emotional aspects of interaction.

  • Animation: the effects of animated diagrams on understanding (H. van Oostendorp)Animated diagrams representing dynamic systems hold great potential for improving the way people learn since there is a direct correspondence between parts of the representation and components of the dynamic system they represent. However, so far the empirical literature failed to find clear benefits of animated diagrams over static pictures - the established way of presenting. In this project we will focus on the conditions where animation leads to improved learning. For instance does focusing attention, by spotlighting the main steps, lead to enhancement of learning?
  • Cognition-based learning principles of games (H. Schijf-Tabachneck, H. van Oostendorp, Ph.D. student). In this AGS project - funded by GATE- we will empirically validate a number of cognitive principles that are important for learning in a (serious) gaming environment. The first issue pertains to determining the right circumstances for building an adequate mental model of a complex task like solving problems in a serious game, by examining which cognitive problems are experienced by users in constructing dynamic mental models and determining how to overcome such problems. We will experimentally investigate a number of factors or ‘enablers’ that should lead to circumventing these problems, particularly regarding presentation rate, focusing attention and knowledge activation. The second issue involves how to maintain the motivation and fun people experience when they solve problems in entertainment games, in “playing” serious games. We hypothesize that one of the ways to induce such fun and motivation is presenting the gamer with informational gaps. The introduction of functional informational gaps in serious games can, under the right circumstances, lead to learning the intended information, and this will be examined in this context. These issues should lead to empirically validated guidelines for construction of a gaming environments meant for learning in an efficient and motivating way.

Within the communication aspect we work on the following projects:

  • Conceptual Discrepancies and Feedback in Communication (R.J. Beun, R. van Eijk and H. Prüst) . The aim of this project is the development of a computational framework for the detection and resolution of conceptual discrepancies in human-computer interaction. In the approach, presuppositions are extracted from the user’s message, expressed in a common vocabulary, and compared with the system’s ontology about the application domain. Discrepancies are detected by the system if it notices type conflicts, particular inconsistencies or ontological gaps. Depending on the kind of discrepancy, the system generates a particular feedback message in order to establish alignment of its private ontology with the mental model of the user. A rudimentary framework has been developed, which is based on Type Theory, that captures the minimal ingredients for the feedback generation process and that provides the basis for further development of concrete feedback rules. This generic framework has been shown to be not only applicable to human-computer interaction but also to human-human and computer-computer communication.

  • Dialogue Protocols and Ontology Alignment (J. van Diggelen (Ph.D student), R.J. Beun, R. van Eijk). Many dialogue protocols developed in the agent communication community build on the assumption that the agents share a common ontology. However, these protocols are difficult to apply in open multi-agent systems and in cases were one of the agents is a human being, since common ontologies are more the exception than the rule. In collaboration with the Intelligent Systems Group a computational framework is developed for ontology alignment by means of dialogue. It starts from the idea that each individual agent is specialized in solving a particular task and the agent requires its own specialized ontology that is, in principle, not directly accessible to other agents. In order to align their ontologies, the agents have to use symbolic interaction, for which it has been shown that the agents need at least some minimal common ground and a way to learn new concepts from each other. We have developed various dialogue protocols for two agents and investigated their characteristics. The framework still has to be tested in real-life scenarios of collaborating personal assistants.

  • Embodied Conversational Agents (R.J. Beun, R. van Eijk). An important development in agent design is the use of so-called Embodied Conversational Agents (ECAs). ECAs are electronic agents that are visually presented in the computer interface with some kind of embodiment – a human, animal or fantasy figure. There is still controversy about the various effects of ECAs on the interaction process. In this project, we focus on the effect of the presence of ECAs in an interface on the retainability of information in relation to the degree to which users anthromorphise these agents. Anthropomorphisation refers to the attribution of human characteristics to the agent. Results of experiments showed that there is a clear positive effect of the presence of an ECA on both memory performance and anthropomorphisation, but no support was found for a correlation of the two concepts. In other words, the memory effects do not depend on the social characteristics ascribed to the ECAs. 

In the knowledge enigineering theme we work on the following projects:

  • Effective knowledge sharing in virtual groups (V. Dignum, E. ten Thij, H. van Oostendorp). Thisresearch aims at deepening the current understanding of the influence of the structure of virtual groups on the knowledge sharing interactions and overall performance of the group. We advocate separating the description of the structure and global behaviour of the group (or organization) from the specification of behaviour of the individual members. That is, global goals determine the roles and interaction norms of group members, which are performed according to the members’ capabilities and aims. Groupstructure and goals as well as individual aims and motivations are determining how trust, reciprocity and delegation are implemented in the group and therefore the success of knowledge sharing in the group. We use multi-agent systems to study these complex processes, because agent systems seem highly qualified to model these processes.
    We will combine simulation methods, based on multi-agent systems, with empirical studies using human groups in a controlled setting.This research will result in a better understanding of the functioning of groups: how to form more effective and efficient virtual groups that incorporate the desires and requirements of their members. Furthermore, we expect that the achieved results can be applied to the design of more effective agent-based systems that support the activity of a virtual group.

  • Trust Issues in Knowledge Sharing (V. Dignum, R. van Eijk, in collaboration with C. Castelfranchi, University of Siena). Knowledge sharing is an essential activity for the management of knowledge. As for any transition, the exchange of knowledge involves risk and trust. Trust has been identified as “the single most important precondition for knowledge exchange”. However, the ways in which trust is built, and the relation of trust to the different methods used for knowledge sharing are not understood well enough. In this project, we aim to formalize knowledge sharing notions such as socialization and articulation, and the influence of trust on knowledge sharing.  Specifically, we aim to develop a model that relates trust to articulation and socialization methods for knowledge sharing.

  • Agent-mediated Knowledge Management (V. Dignum, in collaboration with L. van Elst, DFKI, Germany and A. Abecker, FZI Karlsruhe, Germany). Increasing use of Knowledge Management has been a predominant trend in business in the recent years. Not only is Knowledge Management an important field of application for AI and related techniques, such as CBR technology for intelligent lessons-learned systems, it also provides new challenges to the AI community, like context-aware knowledge delivery. Scaling-up research prototypes to real-world solutions usually requires an application-driven integration of several basic technologies, e.g., ontologies for knowledge sharing and reuse, collaboration support like CSCW (Computer Supported Collaborative Work) systems, and personalized information services. Agent approaches have already been successfully employed for many partial solutions within the overall picture: Agent-based workflow, cooperative information gathering, intelligent information integration, or personal information agents are established techniques in this area. In order to cope with the inherent complexity of a more comprehensive solution, the concept of Agent-mediated Knowledge Management deals with collective aspects of the domain so that the conflicts between desired order and actual behavior in dynamic environments can be resolved.