RICERCA ITALIANA     

ONTOLOGIES, LEARNING OBJECTS AND COMMUNITIES OF PRACTICES: NEW EDUCATIONAL PARADIGMS FOR E-LEARNING

Università degli Studi di Padova

Abstract

The research project aims to fuse the most current and innovative instances, on both a national and international level, concerning e-learning and the semantic web by experimenting their possible application in the field of Education Sciences through the construction of specific “domain ontologies” on the topics of didactic planning, educational communication, assessment and evaluation.
Ontologies are shared knowledge structures, formalized by scientific communities through semantic description languages that can be understood both by people and by software. Initially conceived to resolve problems of formalization, retrieval and sharing of knowledge from the Web, we believe that today their role as conceptual and operative instruments able to favour and support learning processes, is worth studying. This approach should allow, on the one hand the provision of an “ontological” structure which mirrors the conceptual nuclei shared by an academic community, and on the other, adequate management of problems relating to metadating and retrieval of materials and learning objects for teaching.
The idea of a Semantic Web is gaining ground and indeed it aims to go beyond metadating for the efficient searching of documents. This is only possible if the learner can “filter” them through one or more ontological structures which allow him/her to interpret them and place them within a conceptual framework which gives them meaning. Many of these instances are often found in environments for virtual creation and sharing such as Weblogs (or Blogs), whose textual richness can be discovered and enjoyed if the methodology adopted for semantically marking the contents is suitable.
Another important aspect of the research focuses on the fact that communities evolve with time, as do, contemporaneously, conceptual definitions shared by these communities. These definitions provide a “common ground” which consists mainly of meagrely conceptualised and structured tacit knowledge. Virtual communities do not currently have instruments which are advanced enough but also user-friendly for the type of knowledge management which could efficiently resolve this problem which is typical of information and document management. It is precisely to respond to this need that we want to experiment, in virtual communities; instruments which can recognise and construct specified conceptual structures which are shared by these communities, through the fine-tuning and contextual evolution of the EduOnto system, product of the previous PRIN 2003. The final objective however is broader: that is to construct integrated learning environments which are mediated by interpersonal relations. That is we want to attempt the difficult transition from a “formal-ontological” system to a “relational-ontological” system. One of the innovative aspects that the research is founded on is indeed the examination of links and relations between several communities of practice active in different social and discipline contexts according to the theory of “Social Networks”.
Finally the research project cannot go without extension and comparison in an international environment of the Education Science ontologies developed within the project. This objective will be attained through the careful comparative study (in English, French, Spanish and German) of the terminology and concepts referred to in the different knowledge domains present in EduOnto, particularly in the areas of planning, communication and evaluation. <<<

Principal Investigator

Luciano Galliani Università degli Studi di PADOVA

Research Objectives

The research project has been developed in a previous PRIN 2003 on e-learning (“E-learning in university education. Didactic models and pedagogic criteria” coordinated by N. Paparella) that has recently come to an end and entailed the participation of many researchers that are now engaged in the current project. Working upon the definition of a possible pedagogic criteriology, one of the local research units, (Padova's local unit: "Technological platforms, learning modules and representation-research of knowledge” coordinated by Prof. Luciano Galliani) lead to the elaboration of a semantic network about concepts of “Planning”, “Communication”, and “Evaluation”. This part of the project, denominated “EduOnto”, had the immediate aim of building an ontology suitable for describing the actors, processes and technologies of educational sciences, and had as a final result, the setting up of a Learning Object Repository that uses the ontological base for their own “intelligent” consultation.

The aim of the current research therefore is to extend and integrate the achieved results through the construction, experimentation and evolution of an integrated learning environment denominated “EduOntoWiki", that:

- is backed up by an ontological structure relative to educational sciences based upon an active consent of specific communities of practice;
- has a didactic value: through the navigation of ontologies as well as the selection and incorporation of didactic material in the ontological structures (“Learning Object”) supporting learning;
- is realized like a setting where we have a marked formal description (ontology coding) as well as informal (narrativity and contextualization of concepts);
- and in which the possible intra and inter-community relations are explicit and recognized by all the participants through specific social software in order to favour creative solutions to common problems in daily work and study practice

The latter point especially guides our hypothesis according to which only a passage from a technology driven model, (formal-static) to a dynamic and community-driven one, integrated in an open learning environment, could truly render the ontologies built up in the various communities of practice, fruitful and important from a pedagogic and didactic point of view. In this sense the direction that the most promising, current research is taking involves the study of the so-called ‘complex constellations of communities of practice’ (Wenger, 2004), meaning the relations which unite the various communities and make them permeable in such a way that they can reciprocally share knowledge, contextualising and enriching it with new meanings, and in the final analysis, facilitating the solution of complex problems through the recognition of different and little known models, simply because they are perceived as belonging to different environments and disciplines.
The research path which we are proposing therefore, consists in the transformation of the “EduOntoWiki” environment from an ontological/formal system to an ontological/relational one. The Wiki learning environment has been chosen because it allows easy and immediate insertion, modification and sharing of texts and materials: as, for instance, in the case of the “Wikipedia”, an experiment for the creation and up keeping of an encyclopaedia which is open to contributions from everyone. In fact, our initial “vision” conceived the instrument as a formal description of a specific knowledge domain, even if it is mediated by a discussion inside an academic community of practice. The results of the first experimentations lead us to delve into those important reciprocal learning processes that recognize the same learning as a social/relational, intra-communitarian fact (Wenger, 1998), in which the multiple contexts where the learning is situated are precious alternative representations (Lave, 1988). Representations of the knowledge in which the transfer and coding just through rigidly formal modalities (i.e. “ontological”) run the risk of losing all the richness of undertones and natural intentionality typical of narrative (Bruner, 2001). <<<

Timescale

24 months

National and international background

The e-learning paradigm has reached an important epistemological crossroads: there is a need to integrate the current dialogic-informative model, which allows us to interact with people and surf documents on the web, with a model based on the contextual knowledge domains on which we operate: the Semantic Web (Berners-Lee et al., 2001).
The applied research, in particular in the field of education technology, is moving in two directions:
1. the development of solutions for the information exchange, and in general, for the intelligent knowledge management;
2. the development of a collaborative/cooperative approach in knowledge building.

The widespread use of the web in educational contexts is due to the ease with which learning resources can be retrieved and shared: the recent introduction of Learning Objects means that the contents which reside in the different systems used in e-learning platforms need to become re-usable, accessible and interoperable. Each object therefore needs to be described beforehand through unambiguous formalisms (so-called metadata) whose meaning is shared by most communities of possible users, who in this way will be able to retrieve them more easily. If we then assume that the learning process is never confined to the materials used in an online course, but that it is also fruit of the interaction among the members of the group, we must try to provide a support 'scaffolding' (Devedšiæ, 2003) which will facilitate communication and the construction of meaning among all the actors involved (professors, tutors, students) and at the same time represent the structure and contents of the course, whether these are true learning objects or simply teaching modules.
The ontological approach and its intersection with subjects and problems in the wiki world (learning communities, social networks, communities of practice) is developing within this context and according with the computer science utopia/necessity of a semantic web where resources are managed both by systems and as far as contents semantic comprehension is concerned.
An ontology (Guarino 1996) is a formal conceptualisation of a certain field of human knowledge, and is intended as a knowledge base to be shared and re-used in the real world. These ontological structures will, for instance, allow us to surf the documental universe of the Web, no longer through hypertext links from text to text, but from concept to concept; or even to retrieve information in a relevant way without the “noise” that characterizes search engines. In order to achieve this aim, formalized languages have been created (XML, RDF) to mark texts semantically . These languages, which are able to codify knowledge through domain ontologies, can be easily understood both by humans and by ad hoc programs such as semantic browsers (Dzbor et al., 2003) or by specific software agents.
The importance of ontologies has been recognized in different research fields. They are mentioned in relation to a Knowledge Engineering, Knowledge Representation, Knowledge Management and Organization, Database Theory, Information Modelling, Information Integration, Information Retrieval.
Even from an operational point of view the current application areas are different: from medicine to knowledge content standardization, from legal information systems to biological and geographical information systems, from e-commerce to the Natural language processing.
The basis of the ontologies research is about the development of Metamodelling best practices for the creation of semantic models to support it.
The definition of concepts and rules necessary to develop an ontology comes from the reflection of logics and computer science theory on the models and languages for the semantic web.
Even the research line which deals with artificial intelligence is involved in the field of ontologies in order to give: tools for knowledge representation; solutions for sharing and re-using resources; intelligent systems which can work on semantic domain contents, according to specific tasks.
In this line we also find the recent research in the field of Knowledge Management.
Ontological approach and Problem Solving Methods have been integrated in order to complete domain statics knowledge with methods which can be applied to the solution of specific and situated problems.
These methodologies aim at the integration of domain ontologies and process ontologies. The reflection upon meta-ontologies defines another important research trend and an ambitious goal for Knowledge Management. The research of educational technology in the vast field of ontologies is working on the integration of the main advantages in the technological field (referred to a Semantic Web – Artificial Intelligence – Knowledge Management ) with the expectations and demands of the educational community.
As far as the didactic research approaches is concerned, themes emerging are: learning personalization; knowledge negotiated construction in cognitive communities.
Concept-oriented or ontology based architectures seem to be promising solutions in respect with such exigencies. They allow the pertaining retrieval of information in the personalized teaching/learning processes; sharing and re-use of knowledge coded in different Educational Systems according to their goals; development of systems able to adapt their behaviour to the comprehension level and to the interests of the user, delivering and organizing the materials in a personalized and adaptive manner.
The creation of an ontology requires the active consensus of a community of practice. Experts of certain disciplines do not always completely share the same categorizations, interpretations and distinctions. Often this is not only because of the reciprocal irreducibility of fundamental theoretical orders, which is both physiological and necessary, but rather because of the confusion created by the different meanings given to “key” terms in the discipline in question. If it were possible to have an “ontological” reference model which shared lexis and semantics, as regards both terms and their relations, this would probably help to reduce conflicts which arise from misunderstandings and incomprehension. Ontologies created in such a way would also have a significant side-effect for all the actors involved: first of all the definition of a common lexis (Wenger 1998), then a strong push towards the conceptualisation of tacit knowledge, and finally the sharing of a meta-model in which processes, knowledge and relations are shared. Defining ontologies which support educational applications based on the web is therefore no simple task, above all because of the difficulty in formally conceptualising a domain which has always played on the idiosyncratic interpretation of each philosophical/pedagogical approach. An ontology or a series of “educational” ontologies could, however, offer considerable advantages: they could provide a medium which would foster the sharing of knowledge (Learning Object) in a sort of semantic interoperability between different e-learning applications and which would also include pedagogic learning strategies.
We believe, therefore, that the potential effects of the Semantic Web for the world of education and training, and in particular for e-learning, will certainly be positive, but only if governed by a strong pedagogical-methodological reference structure which facilitates integration of the new technological-semantic paradigm into the more recent social theories of learning. <<<