CORDIS provides links to public deliverables and publications of HORIZON projects.
Links to deliverables and publications from FP7 projects, as well as links to some specific result types such as dataset and software, are dynamically retrieved from OpenAIRE .
Deliverables
The e-COGNOS ontology focuses on construction concepts as they relate to e-COGNOS main objective: consistent knowledge representation of construction knowledge items.
The ontology development team has established the following guidelines supporting the development of this ontology:
(1.) The ontology should incorporate already established and recognized industry standard taxonomies and classification systems (namely BS6100, UniClass, bcXML, and Talo 90) and the IFC product model;
(2.) The ontology is NOT intended to be THE ultimate ontology for the BC sector;
(3.) The ontology should be flexible and wide enough to accommodate different business scenarios presented by the end users;
(4.) The ontology must be developed incrementally involving the end users;
(5.) The ontology should be user friendly. i.e. easy to browse and understand;
(6.) The ontology should allow for future expansion;
DARPAs DAML was selected as the language to write the e-COGNOS ontology. The e-COGNOS Ontology Server (e-COSer) provides, through its graphic user interface, a service supporting both the searching of concepts in the ontology and the browsing of the entire ontology, in a user-friendly way.
In lieu of the complicated polymorphic nature of the construction industry, the e-COGNOS ontology has been developed in an iterative fashion to make sure that the resulting ontology is both representative and easy to use. The methodology adopted emphasises two major aspects:
1. Involvement of end users;
2Incremental approach.
The major idea behind this ontology can be summarized in the following sentence: In the context of a Project, a group of Actors uses a set of Resources to produce a set of Products following certain Processes within a work environment (Related Domains) and according to certain conditions (Technical Topics). As such, the proposed taxonomy includes seven major domains to classify these major concepts.
The e-COGNOS ontology has been designed as a collection of coherent sets of concepts in an object oriented fashion. This was needed in lieu of the complexity of the construction domain and, more importantly, to allow for future expansion of the ontology. To show an example, the Bridge Engineering domain is located in Related Domains. It shall include all relevant information about bridge engineering (design practice, codes, etc.). Bridges Domain, however, is a separate domain that is located in the Product Domain. This theme of dividing complicated concepts into several domains is followed throughout the ontology. It allows for greater flexibility in the use of the ontology. For example, a project could include a bridge and a highway. Such project will have only two objects (a bridge and a highway). These two objects could then, if needed, have reference (or pointers) to the extensive domains of Bridge Engineering and Highway Engineering.
IFC presents a very sound product model for the construction industry and as such it is gaining momentum in the industry. The decision to base the e-COGNOS on IFC was made to benefit from its already existing data structures and to allow future linkage to engineering drawings (which could be the source of tremendous knowledge). In contrast to IFC limited number of terms, established classification systems (like BS6100 and UniClass) are rich in the number of terms they include and their coverage of the construction domain. As such, they were also included in the e-COGNOS ontology.
The e-COSer is a Web-based application that plays two complementary roles: ontological support to the KM core services and ontology management. The former, which is indeed the primary function of the e-COSer, consists in taking a list of meaningful keywords extracted from KRs (a query) and producing the respective ranked list of ontological concepts that best match the keywords. The latter supports the management of the ontology, providing services like inclusion/suppression of ontological entities (concepts, attributes, relations), graphical functionalities to browse the ontology, the importing of bcXML-compliant taxonomies, and so on.
The ontology management was not initially included in the functionality of e-COSer but due to problems found when using freeware ontology editors, mainly caused by the fact that DAML+OIL was chosen to represent the e-COGNOS ontology, the e-COSer gradually received management-related functionalities in order to better support the e-COGNOS development.
Main features:
- the eCOSer is a web-based application. The eCOSer functionalities are Web services described in WSDL and accessible via an open SOAP interface;
- Interoperability: the eCOSer interoperates with other e-COGNOS Core Services through the "Web Services" model;
- Ontology-based: the eCOSer provides functionalities to build and manage the e-COGNOS ontology. The eCOSer manages ontological entities, namely concepts, relations, signatures, and instances;
- Scalability: the eCOSer has to be scalable in the sense that the number of requests coming from the other KM Core Services can increase. On the other hand, the e-COGNOS ontology is likely to grow;
- Multi-lingual support: Europe is the main cultural context where e-CKMI would be deployed. In ontological terms, this is reflected by the fact that the ontological concepts have translations in many languages and the eCOSer works in the same way;
-Rely on results/solutions already found: eCOSer has been developed based on previous successful experiences. For instance, the Taxonomy Server developed by eConstruct as part of the eConstruct Reference Architecture was one source of inspiration for eCOSer functionalities.
The e-COGNOS knowledge management infrastructure (e-CKMI) is the main result of the project. Essentially, the e-CKMI is a Web-based solution supported by a domain ontology. It has been developed to operate as a middleware solution in which the concept of service plays the main role.
Broadly speaking, the e-CKMI provides Web-services to support the major functionalities identified in the classical KM cycle, namely: acquisition, cleansing/transformation, indexing, updating, refreshing, searching/discovering, and sharing/dissemination. These services operate supported by an ontology service. Within the e-CKMI, each Knowledge Item (a KI is an individual unit of knowledge) is represented through a set of ontological concepts. This representation or description of a KI is called a Knowledge Representation (KR).
The eCKMI is composed of the following components:
(1) The e-COGNOS User Portal/Manager represents the human interface to the e-CKMI and comprises two applications, namely the Portal and the Manager. The former is used by ordinary users to capture, browse and search knowledge. The latter, used by the e-CKMI administrator, supports the maintenance of the system, including the management of users and their respective access rights, data repositories, ontology-related issues, indexes, etc;
(2) The e-CKMI API describes all the services provided by the "core" e-CKMI to support the requests coming from the client applications. Both e-COGNOS Portal and Manager are considered as client applications regarding the "core" e-CKMI;
(3) The e-COGNOS Kernel plays the role of a "backbone" where the KM services are registered and unregistered;
(4) The KM Services are the ones that actually perform the KM tasks. They were inspired from the "classical" KM cycle and include the following: - The Extractor service actually consolidates acquisition, cleansing/transformation and the extraction itself. As such, it is concerned with collecting knowledge and making it explicitly available through the e-CKMI. In addition, the extractor service provides a "preparation process" which a knowledge item has to pass through in order to become usable in a more effective way. Finally, it provides the ability to extract index terms, which will be used during the searching process;
- The Indexer service closely ties into the extraction service and is concerned with creating indexes to help the later re-use of a knowledge item. The indexes are organised with reference to the ontology. This is accomplished with support of the Ontology service;
- The Profiler service allows a user to create a unique profile which is used both to customise the way in which users view knowledge and provide knowledge in itself via the ability to rank the contents of searches for relevance. Users may also control the way their profiles are going to be seen by others;
- The Searcher service provides functionality to enable a user to look for knowledge items. The searcher looks for KIs which already have the respective KRs stored in the e-CKMI;
- The Discoverer service allows a user to search for KIs that do not have their KRs stored in the e-CKMI. Indeed, the discovery process is intended to be a crawler-based task trying to discover new KIs from local and remote directories and Web sites;
- The Maintenance service is concerned with updating and refreshing. Updating guarantees that the KIs and the ontology can be updated as time progresses. A KI is updated by the user who provided it, whilst the ontology can only be updated by the ontology manager, although all the users may make suggestions concerning improvements to the ontology. Refreshing is intended to cover the consequences of a KI being updated;
- The Ontology service (e-COSer) provides the functionalities required to make the selected ontology available to the other e-COGNOS services, which may require it. It handles all the ontology-related issues within the e-CKMI;
- The Disseminator service is related to capitalising and propagating the acquired knowledge. The e-CKMI supports the creation of workgroups (to which the users can subscribe) and these workgroups may themselves be associated to a set of ontological concepts. The users from a given workgroup may be automatically notified when a KI, which is of interest to the workgroup, is acquired or updated. Of course it is important to emphasise that this knowledge "acquisition-dissemination" cycle is a continuous process.
(5) The e-COGNOS Wrapper is intended to support interoperation between External Services and the e-CKMI;
(6) The External Services represent any specific services provided by a third party application which is not vital to the operation of e-CKMI but may contribute to improve its performance or to extend its functionalities.
Searching for OpenAIRE data...
There was an error trying to search data from OpenAIRE
No results available