From E-Learning Tools to Assistants by Learner Modelling and Adaptive Behavior

Humans need assistance in learning. This is particularly true when learning is supported by modern information and communication technologies. Most current IT systems appear as more or less complex tools. The more ambitious the problems in the application domain are, the more complex are the tools. This is one of the key obstacles to a wider acceptance of technology enhanced learning approaches (e-learning, for short). In computer science, in general, and in e-learning, in particular, we do need a paradigmatic shift from tools of a growing complexity to intelligent assistants to the human user. Computerized assistants that are able to adapt to their human users’ needs and desires need some ability to learn. In e-learning, in particular, they need to learn about the learner and to build an internal model of the learner as a basis of adaptive system behavior. Steps toward assistance in e-learning are systematically illustrated by means of the authors’ e-learning projects and systems eBuT and DaMiT. These steps are summarized in some process model proposed to the e-learning community.

Building Intelligent Multimodal Assistants Based on Logic Programming in the Meme Media Architecture

This chapter introduces a software architecture to build intelligent multimodal assistants. The architecture consists of three basic components: a meme media system, an inference system, and an embodied interface agent system that makes multimodal presentations available to users. In an experimental implementation of the architecture, the author uses three components as the basic framework: Intelligent Pad for a meme media system, Prolog for a logic programming system, and Multimodal Presentation Markup Language (MPML) for controlling an interface agent system. The experimental implementation shows how character agents are defined in a simple declarative manner using logic programming on meme media objects.

Modeling Confidence for Assistant Systems

An intelligent assistant system is supposed to relatively autonomously aid a master in problem solving. The way of doing so that is favored in this chapter involves master-assistant communication that leads to the assistant being exposed to a goal the master wants to achieve. The assistant—if that is possible—then creates problem solution procedures, the implementation of which is supposed to result in the goal being achieved. The assistant then chooses an implementation that can be expected to fit well its master dispositions. An assistant thus needs to know parts of its master’s cognitive structure and to be capable of reasoning about it. The chapter proposes to use models as composite verdictive entities by means of which a master may refer to a domain of individuals. The concept of judgment is assumed to be the simple verdictive entity out of which models are composed. Models are used for representing cognitive states of masters. In particular, model extensions are considered, that is, models that are closed with respect to predicate negation of judgments, conjunction of judgments, and conditional judgments. The concept of confidence is formalized and applied to model extensions for deriving quantitative assertions about judgments. It is briefly discussed how the proposed theory of confidence in judgments can be used in intelligent assistant systems.

Mathematics in Virtual Knowledge Spaces

The workplace of scientists and engineers is about to change: even though numerical software and computer algebra systems remove the burden of routine calculation, it becomes more important than ever to get familiar with new concepts and methods quickly. Given the rapid growth of knowledge in today’s sciences, traditional “learning on supply” (i.e., defining the learning goal as the accumulation of knowledge) is no longer applicable; instead, adequate learning and teaching methods have to be established to guide learners towards efficient self-controlled learning.

Intelligent Support for Building Knowledge Bases for Natural Language Processing

In this chapter we discuss ways of assisting experts to develop complex knowledge bases for a variety of natural language processing tasks. The proposed techniques are embedded into an existing knowledge acquisition framework, KAFTIE, specifically designed for building knowledge bases for natural language processing. Our intelligent agent, the rule suggestion module within KAFTIE, assists the expert by suggesting new rules in order to address incorrect behavior of the current knowledge base. The suggested rules are based on previously entered rules which were “hand-crafted” by the expert. Initial experiments with the new rule suggestion module are very encouraging as they resulted in a more compact knowledge base of comparable quality to a fully hand-crafted knowledge base. At the same time the development time for the more compact knowledge base was considerably reduced.

Interaction Scenarios for Information Extraction

This chapter deals with the question of how the interaction between humans and computers may be designed to enable both partners to bring in their respective strengths. In the context of information extraction from semi-structured documents, several scenarios of interaction are identified and analyzed with respect to their adequacy for cooperative problem solving. The theoretical considerations are illustrated by a particular interaction approach called consistency queries. The chapter aims to initiate a discussion concerning the design and investigation of interaction scenarios which enable humans and machines to collaboratively solve problems.

Wrapper Induction Programs as Information Extraction Assistants

These days, search engines are useful tools relying on quite elaborated technologies which, albeit their enormous frequency of usage and the sophistication of the technologies invoked, continuously frustrate thousands of users. They are simply tools not easy to wield, but they are far from being assistants to their human users. Wrappers are extraction procedures which exploit knowledge of how to wrap information into documents for extracting such information on demand. Wrappers alone are nothing more than highly specialized tools. But the development of wrapper types and information ex-traction technologies has paved the road for a transformation of information extraction tools into intelligent assistants. Such a technology transformation program is outlined and discussed in some detail. The chapter stresses the dichotomy of internal mechanisms and external assistance behavior. It is the authors’ general intention to contribute to the discussion about perspectives of intelligent systems’ assistance. Which properties, features or “systems’ abilities” do establish intelligent assistance beyond the limits of comfortable tools? What are related advantages and drawbacks? What are the potentials and what are the risks of the general trend from tools to assistants? Wrapper induction serves as a representative case study.

Building a Virtual Trainer for an Immersive Haptic Virtual Reality Environment

This chapter describes the virtual trainer we developed for CSIRO’s temporal bone dissection simulator. This simulation software runs on an immersive haptic virtual reality environment. The prototype system uses a task model based on a finite state machine to describe the procedure and interactive landmarks to trace the user’s action in relation to vital structures. This gives the user situation related feedback based on user actions. The future virtual trainer would need to be based on research on intelligent tutoring systems to tailor feedback for the students and maximize their knowledge and skill acquisition. Various features for such a system are described.

Memetic Approach to the Location-Based Ad Hoc Federation of Intelligent Resources

The current Web is a mine of tools and services and the gate to ubiquitous computing environments with a huge number of highly distributed mobile and embedded intelligent resources. Their advanced reuse in assistance systems that support whole processes of our tasks requires federation of resources, that is, both discovery of appropriate resources from a certain scope of resources, and their interoperation and coordination for the required demand. Such a scope and the demand may dynamically change. This chapter focuses on such cases in which both change dynamically, and the change of demand cannot be predicted. Assistance systems for such cases must be able to support us to instantaneously perform federation of Web resources in an ad hoc way. They require new technologies for ad hoc federation of Web resources, and for restricting the scope of discovery in accordance with dynamically changing situations. This paper proposes the use of meme media technologies for ad hoc federation of intelligent resources over the Web. It also proposes the Wiki piazza architecture that works as a repository and lookup service, and combines this service with a location reference service to propose a way of restricting the scope of discovery using location-dependent contexts. These technologies enable location-based ad hoc federation of intelligent resources over the Web and ubiquitous computing environments.

From Planning Tools to Intelligent Assistants

This chapter introduces an approach for creating Web application capable of operating in complex environments is introduced. The approach shows how Meme Media technologies combined with other technologies can be used for solving deferent kinds of problems in particular related to Therapy Planning in clinical trials. Combination of logic programming and fuzzy logic for creating Web applications is also introduced.