NeOn: Lifecycle Support for Networked Ontologies




Скачать 340.61 Kb.
НазваниеNeOn: Lifecycle Support for Networked Ontologies
страница7/9
Дата24.09.2012
Размер340.61 Kb.
ТипДокументы
1   2   3   4   5   6   7   8   9
[PLEASE INSERT SHORT DESCRIPTION] .

Some papers can be downloaded at http://www.dsi.uniroma1.it/~navigli/pubs.html and a webpage can be accessed at http://lcl.di.uniroma1.it/tools.jsp.

TextToOnto

TextToOnto [3] is an ontology learning workbench offering a suite of algorithms that can be combined to build an ontology. In particular, the workbench offers modules for identifying important concepts, learning taxonomic relations and extracting generic binary relations using association rules. The learned ontology is visually presented using a TouchGraph3 based schema visualization technique.

OntoLT

OntoLT [4] is a Protege plugin that allows quick extraction of domain ontologies through a set of mapping rules defined in terms of XPATH queries on XML-based linguistic annotations. Currently the tool comes with two readily defined mappings. The first one allows learning taxonomic relations based on noun phrase information. The second mapping extracts slots. According to this mapping a predicate becomes a slot having as domain its subject and as range its object. Other mappings can be built depending on the needs of the user. The extracted ontology is presented in the Protege environment. Each learned structure (concept or slot) contains detailed information regarding its provenance (i.e., the parts of the corpus where it was identified and the rule which led to its identification).

Text2Onto

Text2Onto [5] is the successor of TextToOnto offering an extended set of ontology learning algorithms. Text2Onto brings several enhancements to its predecessor. First, it relies on a Probabilistic Ontology Model (POM) which allows combining the results of different processing algorithms and representing probabilities for the learned structures. Second, based on this probabilistic model, an improved user interface is built. Learned constructs are listed ordered by their probability of being correct or represented using a graph-based visualisation component. Another advantage of the POM is the traceability of the extraction: each learned structure is stored with a set of pointers to parts of the corpus from where it was extracted. Finally, Text2Onto implements a data driven change discovery mechanism which allows updating the ontology model based on the changes in the underlying data without having to perform the whole extraction from scratch. This grants efficiency and allows the user to trace the evolution of the ontology based on the changes in the underlying data.

OntoGen

OntoGen [7] is a system for semi-automatic topic ontology construction from a collection of documents. Ontology construction is seen as a process where the user is constructing the ontology and taking all the decisions while the computer provides suggestions for the topics (ontology concepts), and assists by automatically assigning documents to the topics, naming the topics, etc. The user can use the suggestions for concepts and their names, further split or refine the concepts, move a concept to another place in the ontology, explore instances of the concepts (in this case documents), etc. The system supports also extreme case where the user can ignore suggestions and manually construct the ontology. All this functionality is available through an interactive GUI-based environment providing ontology visualization and the ability to save the final ontology as RDF. There are two main methodological novelties of the system: (i) suggesting concepts as subsets of documents and (ii) suggesting naming of the concepts. Ontogen can be downloaded at http://ontogen.ijs.si/.

In work of WP2 on collaboration, machine learning and social network analysis methods can be used to analyse collaboration between the users in the networked ontology setting, where the user is working on ontology construction simultaneously with the other users and getting suggestions based on the actions of other users performed while constructing ontology on similar content/topic. In order to perform collaboration analysis, we assume that the data recording the users activities is available containing information on the provided suggestions and the user feedback (eg., user1 got suggestion X that is based on activity of user2 and user1 accepted/rejected the suggestion). In order to perform the collaboration analysis, we will propose and implement methods based on machine learning and social network analysis and connect them to the existing tool for semi-automatic ontology construction OntoGen. Namely, inside the work performed in WP3 on providing context to the user based on the activity of the other users, OntoGen will be extended to support recording and exchange of information between different users on the system. In our case, mapping between the user activities and ontologies they are constructing is based on the assumption that each user has a collection of documents that s/he uses to construct ontology with support of OntoGen. Inside WP2 we will build on the top of that extensions to OntoGen.


References

[1] Buitelaar, P., Cimiano, P., Magnini, B. (2005). Ontology Learning from Text: Methods, Applications and Evaluation, Frontiers in Artificial Intelligence and Applications, IOS Press.

[2] Grobelnik, M., Mladenic, D. (2006). Knowledge discovery for ontology construction. In: Davies, J., Studer, R., Warren, P., (eds.) Semantic web technologies: trends and research in ontology-based systems. Chichester: John Wiley & Sons, cop. pp. 9-27, 2006.

[3] Maedche, A. and Staab, S. (2004). Ontology Learning. In Staab, S. and Studer, R., editors Handbook on Ontologies. International Handbooks on Information Systems. Springer-Verlag., pages 173 – 190.

[4] Buitelaar, P., Olejnik, D., and Sintek, M. (2004b). A Protege Plug-In for Ontology Extraction from Text Based on Linguistic Analysis. In Bussler, C., Davies, J., Fensel, D., and Studer, R., editors, Proceedings of the First European Semantic Web Symposium (ESWS2004), volume 3053 of LNCS. Springer-Verlag, Heraklion, Crete, Greece.

[5] Cimiano, P. and Voelker, J. (2005). Text2Onto - A Framework for Ontology Learning and Data driven Change Discovery. In Proceedings of the 10th International Conference on Applications of Natural Language to Information Systems (NLDB’2005).

[6] Gomez-Perez, A. and Manzano-Mancho, D. (2003). A Survey of Ontology Learning Methods and Techniques. OntoWeb Delieverable 1.5.

[7] Fortuna, B., Mladenic, D., Grobelnik, M. (2006). Semi-automatic construction of topic ontologies, In Knowledge Discovery and Ontologies, Berendt et al. (eds), Springer Lecture Notes.

[8] Mladenic, D., Grobelnik, M. Mapping documents onto web page ontology. In Berendt et al. (eds.), Web mining : from web to semantic web, (Lecture notes in artificial inteligence, Lecture notes in computer science, vol. 3209). Berlin; Heidelberg; New York: Springer, 2004, 77-96.


Recent events focussed on Ontology Learning

Workshop on Ontology Learning at ACL 2006

Workshop on Knowledge Discovery and Ontologies at ECML/PKDD-2005 (KDO-2005)

Workshop on Ontology Learning at ECAI 2004

Workshop on Knowledge Discovery and Ontologies at ECML/PKDD-2004 (KDO-2004)


Possible Directions of Research

All of the current models and tools for ontology learning (see above) are focused on a single user collecting one kind of data on which ontology learning techniques can be applied. This approach to ontology learning may be extended in two points:

Integrating different sources of knowledge: Current tools only evaluate one kind of data at a time (e.g. natural language documents). This may be extended so that the user not only collects documents but also his annotations of the documents (like tagging data) and already existing mini ontologies. The learning system then integrates those informations into one ontology. The goal of this integration of different sources of knowledge is to improve the results compared to taking only one source of knowledge into account.

Integrating the data collections of different users: Current tools do not distinguish between who put a document into the data collection on which the learning techniques should be applied. This has the drawback that it's not possible to integrate data collections of different users if two different but overlapping ontologies should be learned. The goal of the learning tool is then to take the user information into account so that evidence for the intersection of the two target ontologies is collected from the whole data collection and that the differing parts of the ontologies are only learned from the documents of a single user.

4.5 Upgrading database content

Upgrading database content is a functionality which allows a NeOn user (or community of users) to “upgrade” information from an existing database to the Semantic Web by mapping its database schema to an ontology. For details see the following papers: [1], [5], [6], [7] and [8].


Available Models and Tools

The scenario: we have a legacy database and we want to generate semantic web content from it. Until now, the following approaches have been reported in literature:


R2O & ODEMapster

R2O is an extensible, fully declarative and semantically correct language to describe mappings between relational DB schemas and ontologies. R2O is intended to be expressive enough to describe the semantics of these mappings and not only a degree of similarity between entities. R2O is proposed as a DBMS independent high level language that can work with any DBMS implementing the SQL standard. Once mappings are defined in an R2O mapping document, they are processed automatically by the ODEMapster mapping processor to populate an ontology (batch process) or to translate a query (on demand). This is richer than the following ones but its expressiveness is limited to the definition of mappings between database views and ontology concepts and direct mappings between attributes/relations of the ontology and attributes of the database views. See [1].

Another approach, described in [2,3], is based in the semi-automatic generation of an ontology from the database's relational model by applying reverse engineering techniques supervised by the designer. Then mappings are defined between the database and the generated ontology. Because the level of similarity between both models is very high, mappings will be quite direct and complex mapping situations do not usually appear.

A third approach, described in [4], proposes the manual annotation of dynamic web pages which publish DB content, with information about the underlying DB and about how each content item in a page is extracted from the DB. This approach does not deal with complex mapping situations and assumes we want to make our database schema public, which is not always the case.


References

[1] Barrasa J., Corcho O. and Gomez-Perez Asunción. Fundfinder -- a case study of database-to-ontology mapping. In Proc. ISWC Semantic integration workshop, Sanibel Island (FL US), 2003.

[2] Stojanovic L., Stojanovic N. and Volz R. Migrating data-intensive Web Sites into the Semantic Web Symposium on Applied Computing. Madrid, Spain, March 2002

[3] Stojanovic N., Stojanovic L. and Volz R. A Reverse Engineering Approach for Migrating Data-intensive Web Sites to the Semantic Web. Intelligent Information Processing. Montreal 2002

[4] Handschuh S., Staab S. and Volz R. On deep annotation. 12th International World Wide Web Conference, Budapest. May 2003

[5] Barrasa J. Semantic upgrade and publication of legacy data. On book: "Ontologies for Software Engineering and Software Technology". C. Calero, F. Ruiz, M. Piattini (Eds.) October 2006, ISBN-10: 3-540-34517-5, ISBN-13: 978-3-540-34517-6.

[6] Barrasa J., Corcho O., Gómez-Pérez, A. A Semantic Portal for Fund Finding in the EU: Semantic Upgrade, Integration and Publication of heterogeneous legacy data. 10th International Conference on Knowledge-Based & Intelligent Information & Engineering Systems (KES2006). Bournemouth, United Kingdom. October 2006.

[7] Barrasa J., Gómez-Pérez A. Upgrading relational legacy data to the semantic web. 15th international conference on World Wide Web, WWW 2006, Edinburgh, Scotland, UK, May 23-26, 2006. [http://portal.acm.org/citation.cfm?id=1135777.1136019# Poster].

[8] Barrasa J., Corcho O., Gómez-Pérez A. R2O, an Extensible and Semantically Based Database-to-Ontology Mapping Language. Second Workshop on Semantic Web and Databases (SWDB2004). Toronto, Canada. August 2004.

4.6 Collaborative workflow

A collaborative workflow is concerned with the way a community carries out a complex task in a cooperative fashion. In this context, we limit the possible tasks to those involved in ontology-lifecycle management. In order to model a collaborative workflow, we need to consider:

the community and its social aspects;

the modalities by which community members share, negotiate, and exchange knowledge and information;

the technological and computational support that is available for communication and modeling.


When talking of knowledge communities like those targeted by ontology engineering and NeOn, we have to assume a very comprehensive notion of collaboration, including not only ‘standard’ cases in which a number of authors/designers work together towards a shared goal, but also cases in which agents are actively involved who do not share the ultimate goal of a project, or even do not play any ‘active’ role in it (cf. sec. 4.1.3).

As a matter of fact, this happens frequently in ontology engineering, with one user that tries to ‘collaborate’ with the author of some ontology or terminology or design pattern or explanatory text, by conceiving its value, its original intended task, the authoritativeness of its provenance, its fitness to one's own purpose.

That user could start ‘argumenting’ with the author when attempting to reuse the ontology, terminology, pattern, or explanatory text. Argumentation can then be used even in absence of any reply, by manipulating the original knowledge in order to make it available to the current user.

In order to argument and to manipulate knowledge, the user must create a virtual collaboration space that takes into account the context of the author, even if that author will not (or cannot) participate actively in the collaboration.

Based on such a notion of collaboration, WP2 aims at building a suite of models, methods, and tools that support as many as possible collaboration workflows, methods, and strategies, and are open to enrichment and customization.


Available Models and Tools


Community

Communities are modelled as a kind of collective. A formal characterization of the notion of 'collective', based on the DOLCE+DnS ontology [3] and on DDPO (plan ontology, [2]) is given in [1]. .

In its broadest sense, a collective is a collection of agents that is unified by a plan. A plan, in turn, is a description which is conceived by a cognitive agent, defines or uses at least one task and one role (played by agents), and has at least one goal as a proper part. Therefore, collectives are intentional since i) their members are agents; ii) said members play at least one role in a plan; and iii) collectives are unified by plans that are conceived by a(t least one) rational agent.

In a narrower sense, probably more suited to the notion of 'community', a collective is intentional only if it is also partly characterized by one (or more) role(s) defined in its unifying plan. In this case, the plan specifically creates the roles for the collective, as e.g. the different positions inside an organization. Based on this narrower characterization, also other workflow-related notions - such as e.g. 'team' - can be defined (cf. [2]). In either case, a collective is intentional if there is a plan, and if the agentive members of a collection play the roles of that plan.

Typically, each person belongs to more than one collective (or community) at the same time. For example, the same person Peter can have different roles in different plans/workflows, which let different communities emerge. Different roles in the NeOn perspective can be: ‘domain expert for fishery regulations’, ‘discussant’, ‘ontology master’, ‘usage-oriented evaluator’, ‘reengineering expert’, ‘core ontology designer’, etc. The different agents playing roles in a project will constitute a community, but those agents can belong to very different cultural communities, and so on.

For example, the ‘fishery regulations community’ could have been totally unaware of the ‘ontology evaluation community’ until the NeOn project started, and the local functional areas of the related roles become complementary.


Modalities of information management

Peter Kollock [Kol98] outlines a set of design principles to describe cooperation within online communities, merging social studies and practical experiences. He provides a list of requirements, design principles, and best practices for the existence of succesful virtual communities.


Technological and computational support

Although there is a large number of tools that support collaborative workflows, we focus here on tools that support collaborative ontology building workflows (see also T2.3.2 Argumentation).

Compendium: Compendium, a tool developed at the KMI, can be used to support a team for designing and implementing an ontology-based information portal.

IkeWiki: A web-based alternative which explois a bottom-up and relaxed and open contribution philosophy is IkeWiki. IkeWiki is a wiki platform developed at Salzburg research specifically tailored for ontology development. Although It does not provide support to define and configure workflow rules, being a wiki, it lends itself naturally to collaborative work.

Diligent: DILIGENT (DIstributed, Loosely-controlled and evolvInG Engineering of oNTologies) is a methodology for enigineering and evolving ontologies in distributed teams of domain experts. Those experts may propose changes to a shared ontology. For this purpose they make changes to local copies of the shared ontology. A central control board then collects all those locally adapted ontologies and integrates them to a new version of the shared ontology. Central to the methodology is an argumentation model based on the Rethorical Structure Theory (see [5], [6], and secs. 4.2 and 5.7). This argumentation model is used for discussing the design rationale behind changes. Currently, there doesn't exist a specific tool supporting the argumentation model. Instead, standard tools, e.g. for chatting and instant messaging, were used in the DILIGENT case studies in [7].


References

[1] Bottazzi E., Catenacci C., Gangemi A., Lehmann J. (2006), From Collective Intentionality to Intentional Collectives: an Ontological Perspective, Cognitive Systems Research - Special Issue on Cognition and Collective Intentionality, vol.7, 2-3, Elsevier. Available also at http://www.loa-cnr.it/Publications.html.

[2] Gangemi, A., Catenacci, C., Lehmann, J., & Borgo, S. (2004). Task taxonomies for knowledge content: Deliverable of the EU 6FP METOKIS Project D07, http://metokis.salzburgresearch.at.; updated version available at http://www.loa-cnr.it/Papers/D07_v21a.pdf.

[3] Masolo, C., Vieu, L., Bottazzi, E., Catenacci, C., Ferrario, R., Gangemi, A., Guarino, N. (2004), Social Roles and their Descriptions. In D. Dubois, C. Welty, M.A. Williams (eds.), Proceedings of the Ninth International Conference on the Principles of Knowledge Representation and Reasoning (KR2004), Whistler, Canada, June 2-5, pp. 267-277. Available also at http://www.loa-cnr.it/Papers/KR04MasoloC.pdf.

[4] Yiling Lu (2003). B.Sc. Roadmap for Tool Support for Collaborative Ontology Engineering. A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE in the Department of Computer Science, University of Victoria.

[5] C. Tempich, S. Pinto, Y. Sure, S. Staab (2005). Argumentation Ontology for DIstributed, Loosely-controlled and evolvIng Engineering processes of oNTologies (DILIGENT). In: Proc. of ESWC-2005 - European Semantic Web Conference, Heraklion, Crete, Greece, Springer, LNCS, May/June.

[6] W.C. Mann & S.A. (1988). Thompson. Rhetorical Structure Theory: Towards a functional theory of text organization. Text, 8(3): 243-281.

[7] S. Pinto, S. Staab & C. Tempich (2004). Towards a fine-grained methodology for Distributed Loosely-controllled and evolvInG Engineering of oNTologies. In Proceedings of ECAI-2004.

1   2   3   4   5   6   7   8   9

Похожие:

NeOn: Lifecycle Support for Networked Ontologies iconThe Use of uml as a Tool for the Formalisation of Standards and the Design of Ontologies in Agriculture

NeOn: Lifecycle Support for Networked Ontologies iconEnterprise Architecting Lifecycle Management

NeOn: Lifecycle Support for Networked Ontologies icon[edit] Primary lifecycle processes

NeOn: Lifecycle Support for Networked Ontologies iconNetworked Peers For Business

NeOn: Lifecycle Support for Networked Ontologies iconThe importance of trust in the digital networked economy

NeOn: Lifecycle Support for Networked Ontologies iconDans – Data Archiving and Networked Services

NeOn: Lifecycle Support for Networked Ontologies iconVirtual Collaborative Learning Environments for Music: Networked DrumSteps

NeOn: Lifecycle Support for Networked Ontologies iconРуководитель программы: проф. В. А. Иванов Кафедра оптики Научный доц. Ю. Э. Скобло Рецензент: доц. А. А. Пастор Study of the population processes of neon atom

NeOn: Lifecycle Support for Networked Ontologies iconHighly versatile Senior System Administrator, skilled in the architectural design and implementation of high-availability systems for all major networked

NeOn: Lifecycle Support for Networked Ontologies iconNew Media is the term used for networked computerized or digital technologies that permeate society. There are many definitions of New Media, depending on the

Разместите кнопку на своём сайте:
Библиотека


База данных защищена авторским правом ©lib.znate.ru 2014
обратиться к администрации
Библиотека
Главная страница