Adrian Paschke: "Corporate Semantic Web also addresses the pragmatic aspects of using Semantic Web technologies."
26.01.2009
Corporate Semantic Web (CSW) deals with the application of Semantic Web technologies within enterprise settings. Prof. Adrian Paschke, head of the CSW Working Group at the Free University of Berlin, gives a detailed overview of this emerging application area, addresses challenges and opportunities and gives an insight into the economic principles of the corporate use of semantic web technologies. CSW will also be a major topic at this year's I-Semantics Conference from Sept. 2 - 4, 2009 in Graz/Austria and the Semantic Web Meetup in Berlin on March 20, 2009.
In 2008 you founded the Corporate Semantic Web Working Group at the Free University of Berlin. What is its purpose and who are you addressing?
There has been intense research and development in Semantic Web technologies in the last years. Since 2004 there exists a relatively stable core of W3C standards with RDF, RDFS and OWL. Recently, the Semantic Web stack was extended by the RDF query language SPARQL and the W3C Rule Interchange Format (W3C RIF). A large number of tools, software and applications already exist in the public (Semantic) Web and first commercial services are available. Nevertheless, it is still an early technology. There are still gaps in the standards and implementations and the vision of a global machine-understandable Semantic Web will not be fulfilled in the near future due to the high requirements with respect to scalability, security, critical mass of adopters, users, and semantic data on the public Web.
While the vision of a global Semantic Web remains unfulfilled in the next years, the Corporate Semantic Web approach focuses on controlled environments where these issues do not arise. Current Semantic Web tools and standards are already adequate to implement components of such Corporate Semantic Webs. The high potential in many industrial application scenarios and the short-term practicability in closed enterprise settings was the reason to initiate the funded Corporate Semantic Web project and start a new chair addressing this topic at the Free University Berlin. Furthermore, one important long-term goal of Corporate Semantic Web research of the new working group is to widen this focus in the course of time, to develop solutions that scale to a global range, and thereby contribute to the development of the global Semantic Web.
Corporate Semantic Web has two intended meanings for us:
1. (collaborative) workflows, communication and knowledge management based on an infrastructure for enterprise networks which uses Semantic Web technologies.
2. corporate = entrepreneurial usages of Semantic Web technologies
Corporate Semantic Web addresses both the consumer and the produce side, where consumers and producers might be humans as well as automated services, e.g. in business processes and enterprise service networks. This also includes the adequate engineering, modelling, negotiation and controlling of the use of the (meta)data and meaning representations in a (collaborating) community of users or services in enterprise settings where the individual meanings as elements of the internal cognitive structures of the members become attuned to each others’ view in a communicative process. This allows dealing with issues like ambiguity of information and semantic choices, relevance of information, information overload, information hiding and strategic information selection, as well as positive and negative consequences of actions (e.g. in a decision making process).
But, CSW does not only address the technological aspect but also the pragmatic aspect of actually using Semantic Web technologies in enterprises, which includes learning and training aspects as well as economical considerations. Incentives need to be provided to encourage in-house adoption and integration of these new Corporate Semantic Web technologies into the existing IT infrastructures, services and business processes. Decision makers on the operation, tactical and strategic IT management level need to understand the impact of this new technological approach and its adoption costs and return on investment. Therefore, companies will have in mind the economical justifiability of the deployment of new technologies.
In summary, Corporate Semantic Web addresses the enhancement of knowledge creation, management, and use in corporate environments, based on Semantic Web technologies. The Corporate Semantic Web working group
- Identifies and evaluates real-world business use cases
- Offers knowledge exchange through cooperation with business enterprises
- Provides open-source software tools
- Performs economic evaluation of Semantic Web technologies in corporate context
Where do you see the most promising application areas for a corporate semantic web?
There are two major application domains:
1. Automated Semantic Business Processes.
The assumption behind Business Process Management (BPM) is that the uniqueness of an enterprise lies in the way how it manages and executes its business processes. Accordingly, business processes are the most valuable asset of an enterprise. Modern BPM often directly builds upon IT Governance as an strategic instrument of the enterprise business strategy, IT Service Management (ITSM) which describes the change of information technology (IT) towards service and customer orientation, and IT Infrastructure Management (ITIM) which focuses on planning and efficient and effective delivering of IT services and products while meeting quality of service and security requirements. To bridge between this technical IT service view and the business processes, and allow the involvement of humans in these automated processes, Corporate Semantic Web (CSW) technologies provide scalable methods and tools for the machine-readable and human-usable representation of knowledge, especially rules as a means of declaratively describing business rules and IT policies, and ontologies as a means of capturing a domain of discourse such as a business vocabulary which e.g., might be used in modelling business processes or in implementing Semantic Web Services (SWS) for Service Oriented Computing (SOC).
2. Knowledge Management tools.
In particular in the realm of corporate collaboration tools, Semantic Web technologies will support semi-automatic knowledge evolution and dynamic access to and integration of distributed, heterogeneous information sources and knowledge consolidation, e.g. for trend, enterprise structure, and problem recognition. This will enable the mapping from corporate data and human expert information into explicit knowledge and finally into corporate wisdom.
The Corporate Semantic Web project has identified three major areas which are tied together in terms of creating a more competitive business thanks to the semantic web:
Corporate ontology engineering (I) will improve the facilitation of agile ontology engineering to lessen the costs of ontology development and, especially, maintenance. Corporate semantic collaboration (II) focuses the human-centered aspects of knowledge management in corporate contexts. Corporate semantic search (III) is settled on the highest application level of the three research areas and at that point it is a representative for applications working on and with the appropriately represented and delivered background knowledge.
All three parts work together in an integrative Corporate Semantic Web life cycle loop where (1) semantic information is extracted from the existing corporate data (Semantic Search), (2) semantic knowledge such as corporate ontologies or business rules are engineered from this information and semantic-enriched information objects are created (Semantic engineering), and (3) used in collaborative processes and in knowledge-intensive decisions (Semantic Collaboration). These collaborative processes are again semantically analyzed (1) to produce further information bits and in a new loop (1-2-3) are e.g. used to personalize the search and collaboration context.
Enterprises are traditionally rather conservative when it comes to apply new web technologies - best illustrated at Web 2.0. Now suddenly Web 3.0 pops up. What is your strategy to get enterprises interested in these developments?
While technologies and tools that help collaborating and structuring content, such as tagging, wikis, blogs, and collaboration platforms as well as tools for managing and running larger enterprise software and services are in place, companies seek more capable approaches for gaining, managing, and utilizing knowledge required for their adaptive business processes and dynamic enterprise service networks. Often the integration of data and human activities in automated (business) processes / services is purely functional. That is, there is a semantic gap between these two worlds (human / knowledge vs. automated services and processes). Semantic Web technologies allow discovering and transforming existing information into relevant knowledge of practical consequences, trigger automated reactions according to occurred complex events/situations, and derive answers e.g. for decision making to queries from the existing syntactic and semantic information resources. Hence, these technologies are means to bridge between different existing enterprise technologies, Web X.0 technologies as well as between the humans (employees, decision makers etc.) and IT services in enterprises.
I) The CSW project aims at transferring valuable knowledge and results from research into industry. In Berlin there are many small and medium sized enterprises which have an urgent need for Semantic Web technologies to solve their business problems, and which have a need for trained Semantic Web experts. The new CSW working group offers both, knowledge exchange through cooperation with business enterprises, and training of Master and PhD students in the Free University of Berlin as well as training of employees with special learning courses.
II) The industry will not adopt the Semantic Web as stand-alone technologies, but it is looking for generic enterprise solutions, which integrate the Semantic Web technologies in an easy way in their existing IT infrastructures, services and processes. The Corporate Semantic Web project works on such integrated solutions which combine e.g. Web 2.0, enterprise service and Semantic Web technologies, e.g. semantic business rules and semantic business process management solutions or semantic Wikis. Semantic Web enabled enterprise tools are a relevant market.
III) Corporate Semantic Web project develops methods for cost estimation of Semantic Web development processes, ontology / rules use, and ontology / rules maintenance that are adaptable to different corporate environments. Furthermore, methods for evaluating existing ontologies with regard to enterprise relevant usage criteria are needed. Early adopters deploying application oriented solutions for improving their competitive advantages through enhanced knowledge management of semantically rich data will demonstrate incentives for further corporations to follow and thereby accelerate the realization of a global Semantic Web.
From an economic point of view one of the big obstacles to a Semantic Web is the missing value chain for metadata. How do you deal with this problem from a corporate perspective?
I don't think that Semantic Web technologies itself will have a larger market. Some smaller companies might provide stand-alone services around consulting or implementing metadata vocabularies, ontologies / rule bases and stand-alone Semantic Web tools. However, in my opinion the real market value will come from the business solution aggregators which combine the vertical expertise from solution architects <- process architects <- service integrators <- infrastructure providers. Accordingly, the value chain for Semantic Web technologies will be via each of these levels: semantic-enabled enterprise infrastructure -> semantic-enriched business services -> semantic business process -> semantic value-added business strategy -> business solution.
To enable companies to build these value chains and complete the evolution of business processes from sequences of slow-moving, disjointed applications to more responsive semantically integrated end-to-end, situation-aware straight-through flows of action, I see two major steps:
… we need more skilled people in the enterprises at least up to 2050 which can engineer, model and practically use the semantic enriched/enabled enterprise information systems.
… we need general reference models and best practices for selected high value, high impact application domains to offer an additional increase (1) in efficiency, aimed at cheaper and faster delivery of Semantic Web-supported systems, and (2) in reusability of successful solutions exploiting Semantic Web technologies to frequently occurring design problems in different domains. This also includes more standardized domain-specific metadata vocabularies, (business) ontologies and rule bases.
To tackle these issues the CSW group works together with a multidisciplinary team of leading industry stakeholders, research institutes and standardization bodies such as OMG, W3C, EPTS, RuleML, OASIS, ISO, XBRL, MISMO etc. in order to develop reference architectures, models and new ontology / rules standards, e.g. XBRL, MISMO, RIF as well as new online training courses and an international Master program. Recently, Corporate Semantic Web also founded the first Semantic Web Meetup Group in Berlin. The first meeting will be held on March 20th, 2009.
In the future I expect pragmatic/semantic end-user customization tools providing control to information consuming end-users by enabling them to express computationally how to turn existing information into personally relevant information of high practical consequences.
Corporate Semantic Web
www.corporate-semantic-web.de
Berlin Semantic Web Meetup Group
http://www.meetup.com/The-Berlin-Semantic-Web-Meetup-Group/
About Adrian Paschke …
Prof. Dr. Adrian Paschke leads the chair for Corporate Semantic Web at the FU Berlin (FUB) and is vice-director of the Semantics Technologies Institute Berlin (STI Berlin), director of RuleML Inc., and research director of the Centre for Information Technology Transfer (CITT) GmbH. He received his Ph.D. in Information Systems at the Technical University Munich (TUM) with a thesis on Rule-based IT Service Management and Service Level Agreements. His academic carrier has let him to the Ludwig Maximilian University Munich (LMU, German Elite University), the Friedrich Alexander University Erlangen/Nuernberg (FAU), the Technical University Munich (TUM, Germany Elite University), the Technical University Dresden (TUD), and the Free University Berlin (FUB, German Elite University). He was also working as a research staff member at the National Research Council (NRC) in Canada, and the Biotec Innovation Centre Dresden in Germany. Adrian was involved in multiple industrial software development und business engineering projects and has led several knowledge and business/service engineering projects. He is project leader of open-source projects such as the Prova Semantic Web rule engine, Rule Responder, national projects such as the BMBF funded project Corporate Semantic Web, and was involved in several international and EU projects such as the EU Network of Excellence REWERSE, EU Strep Sealife, EU IP DoReMoPat. He is steering-committee chair of the RuleML Web Rule Standardization Initiative, Co-Chair of the Reaction RuleML technical group, founding member of the Event Processing Technology Society (EPTS), voting member of OMG, and invited member of several W3C groups such as the W3C Rule Interchange Format (W3C RIF), where he edits several Semantic Web standards documents.