Exploiting Semantics and User Modeling for Enhanced User Support Liana Razmerita INRIA, Sophia-Antipolis, Project ACACIA 2004 route des Lucioles, 06902 Sophia Antipolis Cedex France
[email protected]
Abstract This paper elaborates on the role of user modeling for enhanced support in Knowledge Management Systems (KMSs). User models, also addressed as user profiles in KMSs, include user’s preferences and are often similar to competency definitions. The proposed user model extends a typical user profile with relevant characteristics of the users such as: level of activity, level of knowledge sharing, type of activity etc. This paper shows that users’ characteristics are key components for enhanced features for KMSs. The paper provides concrete examples of how ontology-based inferences can be used for enhanced user support in retrieving implicit knowledge. The user model is defined as an user ontology based on Information Management System Learner Information Package (IMS LIP) specifications and is integrated in an ontology-based user modeling system.
1
Introduction
Traditionally, knowledge management systems are designed to allow employees to access and utilize the rich sources of data, information and knowledge stored in different forms. “Knowledge Management Systems (KMSs) refer to a class of information systems applied to managing organizational knowledge“(Leidner and Alavi, 2001). From a technological perspective KMSs can also be defined as: “the process of capturing, organizing and retrieving information based on notions like databases, documents, query languages and knowledge mining.” (Thomas and Kellogg, 2001) In the last few years, the vision and goal of knowledge management systems have broadened. Unlike databases, knowledge management systems aim to go beyond the mere administration of data; they need to support: knowledge creation, knowledge sharing, learning processes, collaboration between employees irrespective of their location and a better management of the tacit knowledge. Schutt (2001) emphasizes the fact that the challenge of actual KMSs is to foster knowledge management processes with the final goal to increase the productivity of their employees. Despite the fact that knowledge management systems are more complex and incorporate more functionality; they are not necessarily more successful from a usability perspective. Success drivers of KM initiatives are still a challenging research issue. Different scholars take different viewpoints on what are the success drivers of knowledge management systems: some emphasize the sociological and organizational culture related issues, some focus on the importance of the communities integrated in knowledge management systems, while others just consider the performance of the technology. Prusak (2001) focuses on the user’s satisfaction: “knowledge management shares information management's user perspective; a focus on value as a function of user satisfaction rather than the efficiency of the technology that houses and delivers the information”. In the last few years a user- centred approach has been associated with the design of KMSs. The main research problem addressed in this paper is how user models, models of the knowledge workers, and user modeling processes can be applied in the context of KMSs. An ontology-based user modeling approach is proposed and concrete examples of how ontology-based inferences can be used for expertise modeling are provided. Broadening this research problem, the paper shows the role of user modeling within Knowledge Management System. User modeling mechanisms can be used to determine the level of activity, the level of knowledge sharing and to provide adapted feedback or rewards to the users. Individual motivational drivers or rewards can be correlated to the inferred user behaviour, specific objectives of KMSs (e.g. knowledge sharing, knowledge creation) and change management initiatives. For example, the issue of how to motivate people to share their knowledge is not simply solved by offering people tools for doing this. As resulted from surveys conducted with knowledge workers of a large Spanish organization, the end-users are not intrinsically
motivated to contribute with knowledge or spend time sharing knowledge, especially if it requires extra work. Organizations need to create the enabling factors for the knowledge workers: to be creative, to submit knowledge assets in the system and to diffuse their knowledge. Some incentives for the adoption of knowledge sharing practices might need to be introduced at the whole organizational level. The adoption of KMSs might imply a change process of current work habits of the knowledge workers and implicit changes at the whole organizational level. The paper is organized five sections. The second section introduces some of the challenges associated with the development of a next generation of KMSs. Section 3 provides an overview of user modeling in the context of KMSs. Section 4 presents an ontology-based user modeling and discuss the role of user modeling in an OKMS. Section 5 discusses some evaluation results related to user modelling issues and it indicates work in progress.
2
Trends and challenges for a next generation of Knowledge Management Systems
The complexity of business processes implies that KMSs capture, store and deploy a critical mass of knowledge in various forms. The amount of knowledge available often constitutes an obstacle for finding and retrieving the relevant knowledge. Moreover it is also a factor that contributes to an information overload process for users. Generally, Knowledge Management Systems capitalize explicit knowledge. Explicit knowledge is the most visible form of knowledge and the one we are most familiar with. It is easily written down and includes artifacts and data stored in documents, reports that are available within and outside the organization, software, etc. But, Knowledge Management Systems can, to some extent, address the management of tacit knowledge. Tacit knowledge is more difficult to articulate, and includes the experience, know-how, skills, knacks and the expertise of the people. The paper shows how implicit knowledge can be inferred through ontology-based user modelling processes, as it will be detailed in section 4. Issues that need to be addressed by a next generation of KMSs have been identified by surveying the opinion of the knowledge workers of two big distributed Spanish companies. Perceived needs and limitations of the current knowledge management tools have been emphasized by the knowledge workers in questionnaires. • a need to better organize the content of the KMSs, • a need for enhanced user support for filtering and retrieving the knowledge available in the system, • a need to access the qualifications and experience of peer knowledge workers in the company. The use of ontologies as mechanisms to structure and represent knowledge addresses the first two issues. The problem of user modeling in KMSs relates to the last two issues mentioned above, namely, the information overload issue and the need to better manage tacit knowledge. The need for enhanced user support for filtering and retrieving the knowledge available in the system expressed as “to not get lost” amongst hundreds of documents and to filter “information and noise” relates to research on personalization and adaptive hypermedia. Ontologies, user modelling and service-oriented architectures including: software agents, web services or grid services are emerging technologies to be integrated in KMSs to help handling this complexity. Ontologies are knowledge representation mechanisms for better structuring the domain model. Due to their powerful knowledge representation formalism and associated inference mechanisms, ontology-based systems seem to be a natural choice for a next generation of KMSs. The ontology represents and structures the different knowledge sources in its business domain (Cui et al., 1999; Stojanovic et al., 2001). Existing knowledge sources (documents, reports, videos, etc.) are mapped into the domain ontology and semantically enriched. This semantically enriched information enables better knowledge indexing and searching processes and implicitly a better management of knowledge. The distribution nature of tasks to be handled in a KMS determines a natural choice for the use of web services and multi-agent systems. Societies of agents can act with the purpose of helping the user or solving problems on behalf of the users. Specialized agents are cooperating, negotiating, communicating in order to achieve various functions such as: discovery and classification of new knowledge, search and retrieval of information, the automatic evolution of the domain ontology, etc. Service-oriented architectures are designed to ensure that applications and data can be
accessed by authorized entities regardless of location or technology platform. Such an architecture deliver functionality as a shared service. In order to integrate intelligent/personalized services and to better support their users, information systems need to access or construct and maintain a user model. User models or user profiles contains certain preferences and competency definitions in KMSs. In the following section, a comprehensive view of the user model is presented.
3
User Modeling: problem definition and motivation
Large organizations are more and more concerned with aspects related to how to better manage their human capital and how to allocate resources. Consequently making the experience, the know-how, the knacks of people more visible is a priority for the large organizations. Knowledge workers are the key element in the management of tacit knowledge. The problem of user modelling in KMSs relates to two important issues. On one hand, user modeling processes support the acquisition of competencies, qualifications, work experience explicitly or implicitly. User modeling processes support “expertise discovery”, collaboration and networking as it will be outlined later in the paper. On the other hand, the implicit complexity of KMSs doesn’t necessarily fit the need of the users to have simple systems, systems adapted to their specific needs. In an extended survey related to the vision of the executives on KMSs (Knowings enquete, 2003) keywords such as: utility, simplicity, conviviality, adaptability to the needs and specificity of the enterprise have been emphasized. In this survey personalization is associated with the access to the knowledge assets and with the simplicity of use of the system. The knowledge workers of Indra, a big Spanish company, the end-users of the Ontologging system, suggest “to include mechanisms in order to acquire knowledge about user profile and filter information and noise” and to “adapt the tools to each company or sector”. Creating user models of knowledge workers, provides the basis for personalization and enhanced user support within a KMS. Research on personalization is motivated by the observation that the users of a KMS are different, they have different needs, preferences, and different tasks to handle. The heterogeneity of users, differences in users’ responsibilities, different domains of interests, different competences, and work tasks to be handled in a KMS drives a need to focus on the users, on the user needs and variability in KMS design. Characteristics of the users integrated in the user models are the basis to personalize the interaction with the users. An important strand of research in user modeling aims to enhance the interaction between the users and the systems. The goal of this research is to make complex systems more usable, to speed-up and simplify interactions. (Kay, 2000) Numerous researchers have reported on human-computer interaction issues, human-agent interaction, how to construct adaptive systems, how to tailor and filter information, how to personalize help and dialogue systems and how to personalize interaction in e-commerce and e-learning etc. (Fischer, 2001; Brusilovsky, 2001; Kobsa et al., 2000; Stephanidis, 2001; Kay, 2001; Andre et al., 2000; Fink and Kobsa, 2000, etc.) These traditional application areas of user modeling bring us insights on how user modeling may enhance the users’ interaction with a KMS. A set of adaptation methods and personalization techniques specific to the objectives of KMSs have been described by Razmerita (2004). Amongst these specific objectives are: how to motivate people to create knowledge and to submit new knowledge assets in the system, how to stimulate collaboration and knowledge sharing between knowledge workers irrespective of their location, how to alleviate information overload, how to simplify business processes and work tasks, etc.
4
Ontology-based user modeling
The definition of the user ontology captures rich data about the employee’s profile including characteristics such as: identity, email, address, competencies, cognitive style, preferences, but also a “behavioral profile” as described in the next section. The proposed user model is structured according to Information Management Systems Learner Information Package specifications (IMS LIP). The IMS LIP package (IMS LIP, 2001) is structured in eleven groupings including: Identification, Goal, QCL (Qualifications, Certification and Licenses), Accessibility, Activity, Competence, Interest, Affiliation, Security Key and Relationship. The user model comprises an implicit part, acquired based on the user interaction with the system, and an explicit part, data gathered explicitly through the user profile editor. The user model implemented, as a user ontology using KAON, captures all aspect identified as
relevant for the KMS and for the users. KAON (Karlsruhe Ontology and Semantic Web) framework is a tool suite for managing ontologies (Maedche et al., 2003). As an ontology language, KAON extends RDF/RDFS, so the user ontology is RDF/RDFS compatible. The components of the user model are correlated with the different functions required by a KMS. KMSs need to encourage people to codify their experience, to share their knowledge and to develop an “active” attitude towards using the system. The Behavior concept extends the IMS LIP groupings. The Behavior concept and its subconcepts were introduced to “measure” two processes that are important for the effectiveness of a KMS, namely knowledge sharing and knowledge creation. The Behavior describes characteristics of users interacting with a KMS such as: level_of_activity, type_of_activity, level_of_knowledge_sharing, etc. Based on their activity in the system, namely the number of contributions to the system and the number of the documents read, the user modeling system classifies the users into three stereotypes: readers, writers or lurkers. These categories are properties of the type_of_activity concept. The level_of_activity comprises four attributes that can be associated with the users: very active, active, passive or inactive. The classification of the users according to the type_of_activity or level_of_activity is based on heuristics. For example a lurker is defined as somebody who doesn’t contribute and who reads/accesses very few knowledge assets in the system. Several heuristics are dedicated to capture the interest areas and the level of expertise of the users. The level of adoption of knowledge sharing practices is captured through the level_of_knowledge_sharing. The user states in relation to the level of knowledge sharing are defined as: unaware, aware, interested, trial and adopter using Roger’s terminology related to the attitude of people towards innovation (Angehrn and Nabeth, 1997). Based on the identified characteristics, the system provides feedback, virtual reward or adapted interventions for a behavioural change (e.g. for adoption of knowledge sharing behaviour). By mapping the functions of the KMSs into different domains in which they fit, see table 1, four domains to enhance the functionality of KMSs have been identified. A user model is a key component for: personalization, expertise discovery, networking, collaboration and learning. (Razmerita et al., 2003) These features will be outlined in the following section.
User model
Functions of the KMSs
Identification
Log-on the platform
first_name,
Achieve access to the user’s data through the user profile editor
email,
Search for people/experts
address, etc.
Contact people/experts
Affiliation
Personalize the layout
title
Customize the user interface (adaptation of presentation and modality)
work_unit
Search for experts
Competency/QCL
“Push” content
knowledge interests
Use agent/event-based notification
level of expertise
Personalize the structure Personalize the content
Activity
Share a document
working papers
Manage contributions
projects
Provide incentives for contributors
white papers
Infer the activity of the users in KMSs
documentations prototypes
ProvideResources, submit ressources ProvideMetadata, comment documents, etc.
Behavior
Infer the user’s behavior
Level of activity
Provide incentives for change management
Type of activity
Acknowledge the active users/provide rewards
Level of knowledge sharing
Accessibility
Customize the user interface
language
Personalize the layout
small/ large fonts
Select documents based on the language criteria
likes/ dislikes interface agents
Interest
Organize social activities
hobbies
Provide social networking facilities
Goal
Search/Query Retrieve documents
search for docs/case studies
ProvideResources - Publish
learn, improve skills
ProvideMetadata
search for people
Manage Documents and Folders
contact people
Manage Links Manage Index/Taxonomies/Ontologies Contact people/experts
Table 1 User model components and enhanced features of the KMS
4.1
Personalization
Research on personalization aims to improve the efficiency and the effectiveness of the interaction. Personalization techniques rely on the user’s characteristics captured in user models or user profiles. An important strand of research in user modeling aims to enhance the interaction between the users and the systems. The goal of this research is to make complex systems more usable, to speed-up and simplify interactions (Kay, 2000). The author of the paper distinguishes between a utility function that personalization could bring to the users of a system and a conviviality function with “high touch” impact for the users. From this utility and conviviality perspective various personalized services enable KMSs to adapt their functionality, structure and content to match the needs and preferences of users based on a user model which is stored and updated dynamically (Razmerita, 2004). Fischer (2001) provides some insights in the design of human centered systems supported by user modelling techniques. He emphasizes that high functionality applications must address three problems: (1) the unused functionality must not get in the way; (2) unknown existing functionality must be accessible or delivered at times when it is needed; and (3) commonly used functionality should be not too difficult to be learned, used and remembered. Adaptation methods and personalization techniques relate to specific objectives of KMSs. Amongst these specific objectives are: how to motivate people to create knowledge and submit new knowledge assets in the system, how to stimulate collaboration and knowledge sharing between knowledge workers irrespective of their location, how to alleviate information overload, how to simplify business processes and work tasks, etc (Razmerita, 2003). Personalization techniques rely on the user’s characteristics captured in user models or user profiles. Such personalization mechanisms could include: • direct access to customized relevant knowledge assets; • provide unobtrusive assistance; • helping to find/to recall information needed for a task; • offer to automate certain tasks through implicit or explicit interventions;
Personalisation of a KMS is the process that enables interface customization, adaptations of the functionality, structure, content and modality in order to increase its relevance for its individual users (Razmerita, 2005). Personalization can be achieved in two different ways: based on the agent’s intervention such as information filtering agents or synthetic characters, or based on various types of intelligent services that are transparent for the users also addressed as adaptive techniques in the user modeling literature. The adaptation techniques, at the level of the user interface, can be classified into three categories: adaptation of structure, adaptation of content, adaptation of modality and presentation. For instance, in the range of adaptation of structure, the system can offer personalised views of corporate knowledge based on interest areas and the knowledge of the users, or based on the role and competencies of the users. “Personalised views are a way to organise an electronic workplace for the users who need an access to a reasonably small part of a hyperspace for their everyday work.” (Brusilovsky, 1998) Adaptation of content refers to the process of dynamically tailoring the information that is presented to the different users according to their specific profiles (needs, interests, level of expertise, etc.). The adaptation of content facilitates the process of filtering and retrieval of relevant information. In a KMS, recommender systems, information filtering agents, and collaborative filtering techniques can be applied with the purpose of adaptation of content. The adaptation of presentation empowers the users to choose between different presentation styles, such as different layouts, skins, or fonts. Other preferences can include the presence or absence of anthropomorphic interface agents, the preferred languages, and so forth. Different types of sorting, bookmarks, and shortcuts can also be included in a high functional system. Adaptation of presentation overlaps in a certain extent with interface customisation. The adaptation of modality enables changes from text to other types of media to present the information to the user (text, video, animations, or audio) if they are available in the system. In modern adaptive hypermedia, user can select different types of media. These personalisation mechanisms are described and exemplified with more details in Razmerita (2004). Recently, the concept of contextualization of knowledge goes beyond personalization. In this sense, Dzbor et al. (2004) propose to bring the knowledge to the user through ‘personal portals’ taking into account timely and situational issues and using a wider variety of interaction modalities.
4.2
Learning and Change Management
The role of user models, addressed as student/learner models in the context of learning environments, has been emphasized by the whole body of research dedicated to e-learning, intelligent tutoring systems, interactive learning environments, computer based learning, etc. The student/learner model is often associated with cognitive diagnosis. The student modeling process involves an assessment of what the student knows or/and what the student doesn’t know (knowledge gap, his/her misconceptions) or/and his/her learning style, etc. These individual characteristics are used to tailor the learning processes, to adapt the content of the lessons, to adapt the agents’ interventions to the specific needs of the learner. In the context of KMSs, learning is approached from a change management perspective. In our view learning is not only a process of acquiring new pieces of knowledge but it often involves a behavioral change for the user. Learning is seen as a continuous process, taking place at individual and social level that includes the acquisition of knowledge as well as the contextual use of the knowledge acquired. From this perspective a system can also criticize, provide feedback and stimulus for behavioral change at the individual level. (Angehrn, 1993) A KMS facilitates storing, searching and retrieving of knowledge assets but it also aims at fostering the users’ participation in knowledge sharing and knowledge creation-the adoption of knowledge management behaviors. User modeling processes enable to track the user’s level of knowledge sharing and his/her level of activity, his/her predominant type of activity. By modeling the user’s behavior different types of stimulus agents can intervene or provide feedback to stimulate the user towards learning and change. The adoption of knowledge management behaviors is a learning and a change process which can be facilitated by different intervention strategies acting at cognitive, cultural and social level. On one hand, these inferred characteristics make aware the user about his/her behavior in the system. On the other hand, the identified behaviors can be used to motivate the user to be active in the system: to share and to create knowledge. The system can provide rewards, recognition mechanisms or other
motivational mechanisms. For instance, the system can offer virtual money or it can acknowledge the “knowledge champions” as a mechanism for motivating the users.
4.3
Networking and computer supported collaborative work
The dichotomy of knowledge as tacit and explicit knowledge implies a requisite for sharing knowledge, collaboration and networking. Computer mediated collaboration or computer supported collaborative work has developed a lot in the last years. Different types of communication systems from email to more advanced groupware systems (e.g. shared workspaces, discussion forums, chat systems, instant messaging systems, video-conferenences, etc.) enable virtual interactions, knowledge exchanges, collaboration and learning in distributed working environments. “Collaboration is a social structure in which two or more people interact with each other and in some circumstances, some types of interaction occur that have a positive effect.” (Dillenbourg et al., 1996) Aspects on how to help people collaborate facilitate the exchange of their knowledge enabling learning and thus supporting to achieve individual and collective goals has been the target of innumerable theoretical research and practical projects. (DeSanctis & al. 2001; Gongla and Rizzuto 2001; Lesser and Storck 2001). In the context of learning environments, Dillenbourg and Self (1995) have emphasized that “collaborative style depends on many factors: the learners' characteristics, their relationship, the nature of the task or the context.” For example, the process of grouping people based on domain of interests, roles support building communities of practice or communities of interests. In certain systems like Knowledge Pump or CWall, communities are built based on the user’s domain of interests. (Snowdon and Grasso, 2002) A community of practice is a term introduced by Xerox research, and: “it refers to a group of people who are peers in the execution of real work” (Skyrme, 1999). Greer et al [1998] have shown how user model can be used to support peer help and collaboration in distributed workplace environments. Peer Help System (PHelpS) finds peers-knowledge workers which can help different work related tasks. In certain systems collaboration is aided with matchmakers. Matchmaker agents can establish connections between users with similar domains of interests and expertise. (Chen et al., 1998) The result of the networking is the construction of a social network. The user’s social network represents the relationships of one community member with others and with individuals and communities external to the one considered. Included in the social network are: (1) the personal network composed of friends and acquaintances; (2) the affiliation to sub-communities, and (3) the organizational network, such as the boss, the colleagues, and the work acquaintances. The importance of social networks in innovation diffusion, business processes and economics is very well recognized, some of the studies include (Deroïan 2002, Janssen and Jager 2001)
4.4
Expertise discovery or skill management
The problem of expertise discovery, expert finding, skill mining or intellectual capital management have been widely discussed in the knowledge management literature. Making the competencies, the qualifications and the users’ domains of interests explicit enables location of domain experts and a better skill management. A survey of the different expert finder systems and the associated expertise modeling techniques can be found in [Yimam and Kobsa, 2000]. A number of commercial applications incorporate expert finding capabilities: [e.g., Knowledge ServerTM from Autonomy1, Inc.; KnowledgeMail from Tacit Knowledge Systems2, Organik® from Orbital Software3; Raven from Lotus Development Corp4]. Benjamins et al. (2002) points out the importance of possibility to manage and monitor the skills by the employees. They argue that this enables the calculations of the knowledge gaps between current and desired position and this will likely improve involvement and motivation. This is also a valuable option not only for the knowledge workers who need to complete different job related tasks but also for human resource management units especially for large, distributed organizations. 1 2 3 4
http://www.autonomy.com/ http://www.tacit.com/ http://www.orbitalsw.com/ http://www.lotus.com/home.nsf/welcome/km
Organization
works_at
cooperates_with
User
ganization Project
works_on
name
Figure 1 Application scenario of the user ontology Let’s have a look at a more concrete example of applying ontologies for user modeling and expertise finding in the context of KMSs. It was shown above that the user ontology describes various properties and concepts relevant for the user model. The concepts of the user ontology are bridged with the concepts of the domain ontology through properties. Figure 1, depicts in a graph-based representation a part of the concepts and properties of the user ontology. Concepts are represented with green ovals while properties of the concepts are represented with orange ovals. These properties of the different concepts such as: “works_at”, “works_on”, “cooperates_with” facilitate further inferences. For instance, the fact that: “Smith works_on Kmp project”. The range of the property “works_on” is restricted to concept Project. Furthermore, Kmp is described as a project about: Knowledge Management, Skill Management and Semantic Web. Based on these facts the system automatically infers that Smith is interested or has expertise in: Knowledge Management, Skill Management and Semantic Web. (User, works_on, Project) (Project, related_to, Topic) In our examples the previous RDFS tuples are instantiated as following: (Smith, works_on, KmP) (KmP, related_to, Knowledge Management) (KmP, related_to, Semantic Web) (KmP, related_to, Skill_Management) Using inferences implemented in F-logic (Kifer, 1995) an ontology-based KMS can deduce that Smith might be an expert in Knowledge Management, Ontology and Skill_Management. An Ontobroker (Decker et al., 1999) specific syntax written in F-logic to query all the people working in Knowledge Management looks like: FORALL Y, Z >Z] and Z: Project [related_to->>KnowledgeManagement] Thus without requiring people to constantly update their profiles (their expertise, interests), an ontology-based KMS could facilitate finding the experts or the knowledgeable persons of a domain. This example shows how semanticenriched knowledge assets and associated reasoning mechanisms can be used for inferring implicit knowledge.
5
Conclusions
Designing effective knowledge management systems requires not only a focused view, which is achieved by considering organizational imperatives and technological solutions, but it also benefits from a user-centred perspective that considers the individual needs of the users (e.g. work tasks, responsibilities), individual motivational drivers, usability and ergonomics issues. The evaluation of the system has emphasized that the user satisfaction depends on how useful and how usable is perceived the system, but organizational culture and managerial factors play an important role too. The evaluation has been done combining the questionnaire with other empirical evaluation methods: focus group discussion and semi-structured interviews. The main issues addressed by the evaluation of the user modelling processes were: employees view on sharing personal information and user modelling processes, the perceived need of personalization of KM tools and the use of knowledge distribution agents, knowledge sharing incentives. The analysis of the results has shown that certain users are concerned with privacy and trust issues. This category of users seems to be reluctant related to the use of their data by the organization. Therefore according to the user opinion the user profiles should be made partially available to the other end-users. The behaviour of the users in the system can be associated with incentives provided to the users to share their knowledge and be active in the system. Of course the issue of sharing knowledge and contributing with knowledge to the system is complex and it shouldn’t be limited to simple incentives. It might imply changes of the current work practices and it can be associated with other managerial interventions. We have surveyed different types of incentives a company might use to stimulate knowledge sharing and knowledge creation. According to the user’s opinion from Indra, reputation and promotion in organization would be the right incentives to stimulate a knowledge sharing culture in the organization. However a bonus associated with the salary seems to be also a right incentive for experts to spend extra time-sharing their knowledge. Some expert knowledge workers have expressed their concern in being recognized as experts and having to do extra work. Although a lot of research has been conducted in the area of KMSs and many software platforms have been developed as knowledge managing systems, very little work has been done in the field of user modeling for KMSs. The problem of user modelling relates to two important issues for KMSs: information overload and the need to better manage tacit knowledge. The need for enhanced user support for filtering and retrieving the knowledge available in the system expressed as “to not get lost” amongst hundreds of documents and to filter “information and noise” relates to research on personalization and adaptive hypermedia. The paper has shown that the user model is a key component for providing enhanced features like: personalization, expertise discovery, networking, collaboration and learning. Future work involves the test of the user modelling use case scenario, presented in the previous section, with associated OWL (Ontology Web Language) reasoning mechanisms within a service-oriented architecture. Serviceoriented architectures are designed to ensure that applications and data can be accessed by authorized entities regardless of location or technology platform. Such architectures deliver functionality as a shared service.
Acknowledgment The work reported in this paper was done in the context of the Ontologging project, an EU funded project, at CALT, INSEAD, and it has been finalized during a postdoctoral position within ACACIA project, at INRIA, Sophia Antipolis. Thanks are due to Alain Giboin for feedback on this paper.
References Angehrn, A.& Nabeth, T. (1997). Leveraging Emerging Technologies in Management-Education: Research and Experiences, European Management Journal, Elsevier, 15, pp. 275-285. Benjamins V. R., Cobo, J., M., L., Contreras, J., Casillas J., Blasco J., de Otto B., García J., Blázquez M., Dodero J. M. (2002): Skills Management in Knowledge-Intensive Organizations. EKAW 2002: pp. 80-95 Chen, J., R., Mathe, N. & Wolfe S., (1998). Collaborative Information Agents on the World Wide Web, In ACM DL, pages 279-280. Davenport, T., H., & Prusak L. (1998). Working Knowledge: How Organizations Manage What They Know, Harvard Business School Press.
Dillenbourg, P. & Self, J. (1992). A framework for learner modeling. Interactive Learning Environments, 2 (2), 111137. Deroïan, F. (2002). Formation of social networks and diffusion of innovations, Research Policy, Volume 31, Issue 5, July 2002, Pages 835-846. Dore, L. (2001). Winning through Knowledge: How to Succeed in the Knowledge Economy, Special Report by the Financial World, The Chartered Institute of Bankers in association with Xerox. London. DeSanctis, G., Wright, M. & Jiang, L. (2001). Building a Global Learning Community, CACM, Issue: Global Applications of Collaborative Technology, Vol. 44, No. 12. Dzbor, M., Motta, E.,.Uren, V., Lei, Y. (2004). Reflection on the future of knowledge portals. AIS SIGSEMIS Bulletin 1(2), pp.32-35, July. Greer, J., McCalla, G., Collins, J., Kumar, V., Meagher, P., and Vassileva, J. (1998). Supporting Peer Help and Collaboration in Distributed Workplace Environments. International Journal of Artificial Intelligence in Education 9, 1998, 159-177. IMS LIP, (2001) IMS Learner Information Package http://www.imsproject.org/aboutims.html, 2001 Janssen, M.,A., and Jager, W., (2001),Fashions, habits and changing preferences: Simulation of psychological factors affecting market dynamics, Journal of Economic Psychology, Volume 22, Issue 6, December, Pages 745772. Kay, J, (2000). User modeling for adaptation, in User Interfaces for All, Stephanidis (ed), C, Salvendy, G, (General Editor), Human Factors Series, Lawrence Erlbaum Associates, 271—294. Leidner, D., Alavi, M., (2001). Review: knowledge management and knowledge management systems: conceptual foundations and research, INSEAD-MIS Quarterly, vol. 25 (no. 1), pp. 107-136, 2001. Maedche, A., Motik, B., Stojanovic, L., Studer, R. & Volz, R. (2002).Ontologies for Enterprise Knowledge Management, IEEE Intelligent Systems, November/December. Nonaka, I. & Hirotaka T., (1995). The Knowledge-Creating Company, Oxford University Press, Razmerita, L., Angehrn A. & Nabeth, T., (2003). On the role of user models and user modeling in Knowledge Management Systems, Volume 2 of the Proceedings of HCI International, Greece, pp. 450-456. Razmerita, L, (2003). “User Model and User Modeling in Knowledge Management Systems: An Ontology-based Approach”, PhD thesis, University of Toulouse, France Razmerita, L. (2004.) User modeling and personalization of the Knowledge Management Systems, in Adaptable and Adaptive Hypermedia, edited by Sherry Chen and George Magoulas, published by Idea Group Publishing. Snowdon, D. & Grasso, A., (2002). Diffusing information in organizational settings: learning from experience, Conference on Human Factors and Computing Systems, Minnesota, pp. 331 – 338. Stojanovic, L., Maedche, A., Motik, B. & Stojanovic, N. (2002). User-Driven Ontology Evolution Management, Proceedings of the 13th European Conference on Knowledge Engineering and Management, EKAW-2002, Springer, LNAI, Madrid, Spain. Skyrme, J. (1999). Knowledge Networking, Creating the Collaborative Enterprise, Butterworth-Heinemann
