On the Move to Meaningful Internet Systems 2004: CoopIS, DOA, and ODBASE

We would like to welcome you to the Proceedings of the Twelfth International Conference on Cooperative Information Systems (CoopIS 04). As in previous years, CoopIS is part of the federated conference On the Move (OTM) to Meaningful Internet Systems and Ubiquitous Computing, this year taking place in Agia Napa, Cyprus, together with the International Symposium on Distributed Objects and Applications (DOA) and the International Conference on Ontologies, Databases and Applications of Semantics for large-scale Information Systems (ODBASE).

Recently, we have seen increasing adoption of business process automation technologies (and emerging standards for business process orchestration) by enterprises, as a means of improving the efficiency and quality of their internal operations, as well as their interactions with other enterprises and customers as they engage in e-business transactions. The next phase of evolution is the rise of the intelligent enterprise, which is characterized by being able to adapt its business processes quickly to changes in its operating environment. The intelligent enterprise monitors its business processes and the surrounding environment, mines the data it collects about the processes to understand how it is meeting its business objectives, and it acts to control and optimize its operations to meet those business objectives. Decisions are made quickly and accurately to modify business processes on the fly, dynamically allocate resources, prioritize work, or select the best service providers. This talk will describe challenges in managing and optimizing the business processes of an intelligent enterprise. We will describe technology approaches that we are pursuing at HP Labs., the progress we have made, and some open research questions.

Previous workflow mining works have concentrated their efforts on process behavioral aspects. Although powerful, these proposals are found lacking in functionalities and performance when used to discover transactional workflow that cannot be seen at the level of behavioral aspects of workflow. Their limitations mainly come from their incapacity to discover the transactional dependencies between process activities, or activities transactional properties. In this paper, we describe mining techniques, which are able to discover a workflow model, and to improve its transactional behavior from event logs. We propose an algorithm to discover workflow patterns and workflow termination states (WTS). Then based on the discovered control flow and set of termination states, we use a set of rules to mine the workflow transactional behavior.

We consider distributed information systems that are open, dynamic and provide access to large numbers of distributed, heterogeneous, autonomous information sources. Most of the work in data mediator systems has dealt with the problem of finding relevant information providers for a request. However, finding relevant requests for information providers is another important side of the mediation problem which has not received much attention. In this paper, we address these two sides of the problem with a flexible mediation process. Once the qualified information providers are identified, our process allows them to express their request interests via a bidding mechanism. It also requires to set up a requisition policy, because a request must always be answered if there are qualified providers. This work does not concern pure market mechanisms because we counter-balance the providers’ bids by considering their quality wrt a request. We validated our process on a set of simulations. The results show that the mediation process supports the providers in adequacy with the user expectations, even if they are sometimes imposed.

Exception handling is a fundamental functionality of workflow management systems (WfMS). User involvement in exception handling is recognized as critical in various situations due to the unpredictability nature of the exceptions that can occur in a running workflow (WF) engine. The problem however is how to orchestrate human ad hoc interventions with a minimum impact on system integrity. The control flow and data integrity dimensions of the impact are analyzed. Our perspective is to allow the maximum latitude possible to user interventions while keeping system correctness. We propose a solution that uses a WF to guide users handling WF exceptions. Furthermore, we extended the WF engine with a propagation mechanism allowing users to involve multiple members of the organization in the exception handling WF. This solution is implemented in the OpenSymphony (OS) platform. The implementation details of the proposed solution in the OS platform are also given in the paper.

Schema matching is a key operation in meta-information applications. In this paper, we propose a new efficient schema matching algorithm to find both direct element correspondences and indirect element correspondences. Our algorithm sufficiently exploits semantic, structure and instance information of two schemas. It has advantages of various kinds of algorithms and hence a learning methodism.

In ebXML the choreography of a business process should be modeled by UMM (UN/CEFACT Modeling Methodology) and is finally expressed in BPSS (Business Process Specification Schema). Our analysis of UMM and BPSS by workflow patterns shows that their expression power is not always equivalent. We use the workflow patterns to specify the transformation from UMM to BPSS where possible. Furthermore, the workflow patterns help to show the limitations of UMM and BPSS and to propose improvements.

The notion of business process is becoming increasingly important in all business and information/communication technology related disciplines, and therefore gets a lot of attention. Consequently, there is a variety of definitions as well as a variety in preciseness of definition. The research reported in this paper aims at clarifying the different understandings and unifying them in a common conceptual framework. Three fundamental questions concerning business processes are investigated: about the difference between business process and information process, about the distinction between the ‘deep’ structure and the ‘surface’ structure of a business process, and about the difference between system and process. These questions are discussed within the framework of the Ψ-theory and the DEMO methodology.

Adaptive process–aware information systems must be able to support ad–hoc changes of single process instances as well as schema modifications at the process type level and their propagation to a collection of related process instances. So far these two kinds of (dynamic) process changes have been mainly considered in an isolated fashion. Especially for long-running processes, however, it must be possible to adequately handle the interplay between type and instance changes as well. One challenge in this context is to determine whether concurrent process type and process instance changes have the same or overlapping effects on the original process schema or not. Information about the degree of overlap is needed, for example, to determine whether and – if yes – how a process type change can be propagated to individually modified process instances as well. This paper provides a formal framework for dealing with overlapping and disjoint process changes and presents adequate migration strategies depending on the particular degree of overlap. In order to obtain a canonical representation of changes an algorithm is introduced which purges change logs from noisy information. Finally, a powerful proof-of-concept prototype exists.

We present a novel transformation method that allows us to map unstructured cyclic business process models to functionally equivalent workflow specifications that support structured cycles only. Our solution is based on a continuation semantics, which we developed for the graphical representation of a process model. By using a rule-based transformation method originally developed in compiler theory, we can untangle the unstructured flow while solving a set of abstract continuation equations. The generated workflow code can be optimized by controlling the order in which the transformation rules are applied.We then present an implementation of the transformation method that directly manipulates an object-oriented model of the Business Process Execution Language for Web Services BPEL4WS. The implementation maps abstract continuation equations to the BPEL4WS control-flow graph. The transformation rules manipulate the links in the graph such that all cycles are removed and replaced by equivalent structured activities. A byproduct of this work is that, if a continuation semantics is adopted for BPEL4WS, its restriction to acyclic links can be dropped.

More and more companies use ”process aware” information systems to make their business processes more efficient. To do this, workflow definitions must be formulated in a formal specification language, as they represent executable derivates of business process descriptions. Both for the less formal descriptions of business processes as well as the workflow definitions, Petri-net based approaches are used successfully. In the literature the business process descriptions are required to be well-structured, leading to a sound workflow definition. We argue that in many cases well-structuredness is too restrictive for practitioners. Relaxed soundness has been introduced previously as a more suitable requirement. The paper presents how methods from controller synthesis for Petri nets can be used to automatically make this type of models sound. For this reason we adopt the idea of controllability for Petri net workflow models.

In distributed collaborative systems, replicated objects, shared by users, are subject to concurrency constraints. All methods [4, 13, 18, 15, 16, 19, 22] proposed to serialize concurrent operations and achieve copies convergence of replicated objects are based on the use of Operational Transformations. In this context, giving the user the ability to undo an operation has been recognized as a difficult problem [1, 2, 3, 12, 14, 20, 21]. The few general propositions to solve the problem sometimes compromise copies convergence and/or users’ intention, insofar as the Operational Transformations used are unsuitable for undo. This paper has a twofold objective. Firstly, it aims to highlight two general conditions (named C3 and C4) that need to be satisfied by any transformation adapted to undo. Secondly, it presents a general undo algorithm based on the definition of a generic undo-fitted transformation, which automatically verifies these conditions. The interest of the proposed method is that the undoing of an operation obeys to the same processing as the one used for regular operations in collaborative systems such as [15,19].

We consider the use of a cluster system for managing autonomous databases. In order to improve the performance of read-only queries, we strive to exploit user requirements on replica freshness. Assuming mono-master lazy replication, we propose a freshness model to help specifying the required freshness level for queries. We propose an algorithm to optimize the routing of queries on slave nodes based on the freshness requirements. Our approach uses non intrusive techniques that preserve application and database autonomy. We provide an experimental validation based on our prototype Refresco. The results show that freshness control can help increase query throughput significantly. They also show significant improvement when freshness requirements are specified at the relation level rather than at the database level.

Data production systems are generally very large, distributed and complex systems used for creating advanced (mainly statistical) reports. Typically, data is gathered periodically and then subsequently aggregated and separated during numerous production steps. These production steps are arranged in a specific sequence (workflow or production chain), and can be located worldwide. Today, a need for improving and automating methods of supervision for data production systems has been recognised. Supervision in this context entails planning, monitoring and controlling data production. Since there are usually alternate solutions, it makes good sense to consider several approaches. The two most significant approaches are introduced here for improving this supervision, the ‘closely coupled -’ and the ‘loosely coupled approach’. In either situation, dates, costs, resources, and system health information is made available to management, production operators and administrators to support a timely and smooth production of periodic data. Both approaches are theoretically described and compared. The main finding is that both are useful, but in different cases. The main advantages of the ‘closely coupled approach’ are the large production optimisation potential and a production overview in the form of a job execution plan, whereas the ‘loosely coupled approach’ mainly supports unhindered job execution without adapting legacy components and offers a sophisticated production overview in form of a milestone schedule. Ideas for further research include investigation of other potential approaches and theoretical and practical comparison.

In this paper we propose a semi-automatic technique for deriving similarities between XML sub-schemas. The proposed technique is specific for XML, almost automatic and light. It consists of two phases: the former one selects the most promising pairs of sub-schemas; the latter one examines them and returns only the similar ones. In the paper we discuss some possible applications that can benefit of derived sub-schema similarities and we illustrate some experiments we have conducted for testing the validity of our approach. Finally, a comparison of the proposed approach with some related ones already presented in the literature, as well as a real example case aiming at better clarifying it, are presented.

Querying and integrating sources of structured data from the Web in most cases requires similarity-based concepts to deal with data level conflicts. This is due to the often erroneous and imprecise nature of the data and diverging conventions for their representation. On the other hand, Web databases offer only limited interfaces and almost no support for similarity queries. The approach presented in this paper maps string similarity predicates to standard predicates like substring and keyword search as offered by many of the mentioned systems. To minimize the local processing costs and the required network traffic, the mapping uses materialized information on the selectivity of string samples such as q-samples, substrings, and keywords. Based on the predicate mapping similarity selections and joins are described and the quality and required effort of the operations is evaluated experimentally.

Recovery in agent systems is an important and complex problem. This paper describes an approach to improving the robustness of an agent system by augmenting its failure-handling capabilities. The approach is based on the concept of semantic compensation: “cleaning up” failed or canceled tasks can help agents behave more robustly and predictably at both an individual and system level. However, in complex and dynamic domains it is difficult to define useful specific compensations ahead of time. This paper presents an approach to defining semantic compensations abstractly, then implementing them in a situation-specific manner at time of failure. The paper describes a methodology for decoupling failure-handling from normative agent logic so that the semantic compensation knowledge can be applied in a predictable and consistent way– with respect to both individual agent reaction to failure, and handling failure-related interactions between agents– without requiring the agent application designer to implement the details of the failure-handling model. In particular, in a multi-agent system, robust handling of compensations for delegated tasks requires flexible protocols to support management of compensation-related activities. The ability to decouple the failure-handling conversations allows these protocols to be developed independently of the agent application logic.

This paper describes PUMAS, our proposal of architecture which uses ubiquitous agents in order to give to nomadic users (users who often change of location) access to a Web Information System (WIS) through their Mobile Devices (MDs). PUMAS focuses on the exchange of information between MDs seen as peer ubiquitous agents and takes into account the user’s needs, location and the limited configuration of her/his MD for displaying the information. PUMAS also handles two important aspects of a nomadic use of a WIS: user location (provided by a GPS device) and information distribution among several heterogeneous MDs (controlled by a Peer to Peer approach). In addition, we present the Agent Class Diagram and the Interaction Diagrams of PUMAS in the Agent UML (AUML) notation [8]. An application relying on PUMAS is described.

The integration of event information from diverse event notification sources is, as with meta-searching over heterogeneous search engines, a challenging task. Due to the complexity of event filter languages, known solutions for heterogeneous searching cannot be applied for event notificationIn this paper, we propose the concept and design of a Meta Service for Event Notification. We define transformation rules for exchanging event filter definitions and event notifications between various event services and sources. We transform each filter defined at a meta-service into a filter expressed in the language of each event notification source. Due to unavoidable asymmetry in the semantics of different langues, some superfluous information may be delivered to the meta-service. These notifications are then post-processed to reduce the number of spurious messages. We present a survey and classification of filter languages for event notification, which serves as basis for the transformation rules. The proposed rules are implemented in a prototype transformation module for a Meta Service for Event Notification.

Publish/subscribe middleware provides efficient support for loosely coupled communication in distributed systems. A number of different distributed message-filtering algorithms have been proposed. So far, a systematic comparison and analysis of these filter algorithms is still missing.This paper proposes a classification scheme for distributed filter algorithms that supports the theoretical and practical analysis of these algorithms. We present a first cut theoretical evaluation and a subsequent practical evaluation of promising candidate algorithms. Factors that are considered include the characteristics of the underlying network and application-related constraints.Based on the findings of these evaluations, we conclude with a summary of the strengths and weaknesses of the algorithms that we have studied.

This paper presents a collaborative model for agricultural supply chains that supports negotiation, renegotiation, coordination and documentation mechanisms, adapted to situations found in this kind of supply chain – such as return flows and composite regulations. This model comprises basic building blocks and elements to support a chain’s dynamic execution. The model is supported by an architecture where chain elements are mapped to Web Services and their dynamics to service orchestration. Model and architecture are motivated by a real case study, for dairy supply chains.

Content-Addressable Networks (CAN) manage huge sets of (key, value)-pairs and cope with very high workloads. They follow the peer-to-peer paradigm: They consist of nodes that are autonomous. This means that peers may be uncooperative, i.e., not carrying out their share of the work while trying to benefit from the network. This article deals with this kind of adverse behavior in CAN, e.g., not answering queries and not forwarding messages. It is challenging to design a forwarding protocol for large CAN of more than 100,000 nodes that bypasses and excludes uncooperative nodes. We have designed such a protocol, with the following characteristics: It establishes logical networks of peers within the CAN. Nodes give positive feedback on peers that have performed useful work. Feedback is distributed in a swarm-like fashion and expires after a certain period of time. In extreme situations, the CAN asks nodes to perform a proof of work. Results of experiments with 100,000 peers are positive: In particular, cooperative peers fare significantly better than uncooperative ones.

Collaborative layouting occurs when a group of people simultaneously defines the layout of a document at the same time into a coherent set of meaningful styles. This activity is characterized by emergence, where the participants’ shared understanding develops gradually as they interact with each other and the source material. Our goal is to support collaborative layouting in a distributed environment. To achieve this, we first observed how face-to-face groups perform collaborative layouting in a particular work context. We report about the design and evaluation of a system which provides a large workspace and several objects that encourage emergence in collaboration conflicts. People edit documents that contain the raw text and they enhance the readability by layouting this content.

Much work has been done on building content-based publish/subscribe systems over structured P2P networks, so that the two technologies can be combined together to better support large-scale and highly dynamic systems. However, existing content-based routing protocols can only provide weak reliability guarantee over structured P2P networks. We designed a new type of content-based routing protocol over structured P2P networks – Identifier Range Based Routing (IRBR) protocol, which organizes subscriptions on the basis of the identifier range of subscribers. It provides strong reliability guarantee and is more efficient in event delivery. Experimental results demonstrate the routing efficiency of the protocol.

One of the first areas where virtual reality found a practical application was military training. Two fairly obvious reasons have driven the army to explore and employ this kind of technique in their training; to reduce exposure to hazards and to increase stealth. Many aspects of military operations are very hazardous, and they become even more dangerous if the soldier seeks to improve his performance. Intelligent Virtual Agents (IVAs) are used to simulate a wide variety of high fidelity simulation scenarios like the one we have described above. The work described in this paper focuses on military humanitarian assistance and disaster relief in Co-operative Information System (CIS), emphasising on how important it is for IVAs inhabiting this kind of scenarios to be aware of their surrounding before interacting with it. We also highlight the importance of increasing the psychological “coherence” between the real life and the virtual environment experience in order to reflect human perception, behaviour and reactions.

Companies are more and more focusing on their core competencies and therefore are increasingly collaborating with partners, forming value networks in order to better react to fast changing market requirements. Value networks for complex products and services are exposed to dynamic changes making it very difficult for a requestor of a product or service to keep track on all suppliers or service providers contributing to a specific request. The modelling of such networks becomes very complex and therefore often diminishes the practical impact of supply network management approaches. This paper proposes a concept for the dynamic modelling of supply networks, introducing the concept of self modelling demand driven value networks. The concept has been proven by applying it to the business domain of strategic supply network development and by implementing a prototype application for the domain mentioned. Apart from introducing the concept, technical issues and design aspects of the implementation are discussed in this paper and the prototype is introduced.

Project e-Sharing has developed an e-marketplace that supports the efficient sharing of resources among companies of the constructions sector (primarily) according to their time-varying needs. In this paper, we present the e-Sharing Trader system, which supports the leasing of resources by means of electronic auctions and negotiations. The Trader auction-related part supports a wide variety of single- and multi-object auctions together with innovative bidding agents for the English and the ascending clock auctions; these agents place bids on behalf of the users according to their specified preferences. The Trader negotiation-related part supports direct multi-attribute negotiations between users by means of a semi-structured negotiation protocol, automated agent-aided price negotiation, and two-object multi-attribute negotiations so that a user leases either two complementary resources or none, or exactly one out of two substitute resources. We also compare the e-Sharing Trader with existing e-marketplaces and discuss the advantages of our work.

Changes to engineering products need to be effectively co-ordinated with other members in the development team. Current approaches to managing change have limitations and often require further manual, ad-hoc activities to co-ordinate each change. This results in an inconsistent approach to change management. Our approach uses change processes to manage changes to products. Change processes are modelled using UML Activity Diagrams, which clearly show how human and technical activities are co-ordinated for each type of change operation. Change process enactment has been achieved in a research prototype by integrating workflow technology with a development system that supports versioning. This process-based approach to change management provides standardised, auditable change processes, which support change throughout the product development lifecycle. This negates the need for manual, ad-hoc activities to co-ordinate changes, resulting in a more consistent approach to managing change.

eHome systems are essentially component-based systems. One of the main reasons preventing a wide application of eHome systems in practice is the effort needed to interconnect all appliances, necessary controller and infrastructure components to benefit from derived value-added services. In the area of software engineering, this problem is addressed by configuration management and software deployment. In this paper, we introduce a language, which forms a basis for describing coarse-grained and abstract scenario defaults up to complete deployment information to carry out the actual installation in an eHome. We present a tool supporting the automatic transformation of an abstract input document into a complete deployment document adapted to a specific eHome environment. The tool is based on the description language.

This paper presents a proposal of a context-based architecture to achieve the required synergy among the ubiquitous computing devices of an intelligent environment. These devices produce context information that models the behaviour of the environment. This context information is the glue among the devices and the context-aware applications. The generated context information provides a common view of the world. A blackboard architecture allows to share this context information and a context model is proposed to represent it. A prototype of such a smart room has been developed, including several devices as well as a set of context-aware demonstrators. They work together employing the context information stored on the blackboard.

Recommender Systems allow people to find the resources they need by making use of the experiences and opinions of their nearest neighbours. Costly annotations by experts are replaced by a distributed process where the users take the initiative. While the collaborative approach enables the collection of a vast amount of data, a new issue arises: the quality assessment. The elicitation of trust values among users, termed “web of trust”, allows a twofold enhancement of Recommender Systems. Firstly, the filtering process can be informed by the reputation of users which can be computed by propagating trust. Secondly, the trust metrics can help to solve a problem associated with the usual method of similarity assessment, its reduced computability. An empirical evaluation on Epinions.com dataset shows that trust propagation can increase the coverage of Recommender Systems while preserving the quality of predictions. The greatest improvements are achieved for users who provided few ratings.

Information systems must establish trust to cooperate effectively in open environments. We are developing an agent-based approach for establishing trust, where information systems are modeled as agents that provide and consume services. Agents can help each other find trustworthy parties by providing referrals to those that they trust. We propose a graph-based representation of services for modeling the trustworthiness of agents. This representation captures natural relationships among service domains and provides a simple means to accommodate the accrual of trust placed in a given party. When interpreted as a lattice, it enables less important services (e.g., low-value transactions) to be used as gates to more important services (e.g., high-value transactions). We first show that, where applicable, this approach yields superior efficiency (needs fewer messages) and effectiveness (finds more providers) than a vector representation that does not capture the relationships between services. Next, we study trade-offs between various factors that affect the performance of this approach.

In the business world, business coordinations are becoming global, execution of multi-party contracts has to be considered a vital point for successful business coordinations. The need for a new multi-party contract model is thus becoming evident. However, there is little known on how to formally model a multi-party contract. In this paper, we investigate how a contract involving multilateral parties can be modeled more easily for finding the contract responsible for given contract violations.

In a mobile environment, location management is fundamental in supporting location-dependent applications. It is crucial to reduce the communication overhead in location management, due significantly to the costly uplink traffic for mobile hosts reporting their location to the server. To reduce uplink traffic, the group-based location management scheme exploits the spatial locality of mobile hosts to generate an aggregated location update from a group leader for group members agglomerated through a dynamic clustering algorithm. Due to the mobility of group members, a leader may be decoupled from a group voluntarily or involuntarily. An intuitive approach to address leader departure is to re-execute the clustering algorithm among leaderless group members. However, system performance may suffer, due to the absence of a group leader for a period. In this paper, a leadership maintenance scheme is designed based on the notion of a secondary leader, which is ready for assuming the role of a primary leader. The turnover activation policy identifies endangered primary leader and triggers the turnover procedure, which involves host interaction in leadership handover from the primary to secondary leader. Simulation study shows that our leadership maintenance scheme is effective to further reduce the costly uplink traffic and aggregated cost in the group-based location management scheme.

The P2P paradigm is increasingly receiving attention in many contexts such as Cooperative Information Systems. In this paper we present a P2P lookup service based on a hash table distributed on a hierarchical data structure (a forest). The novelty of our proposal is that it provides a dynamically adapting (to the number of peers) routing load biasing for decreasing the cost of peer insertion and deletion w.r.t. the state of the art. This makes our system particularly suited to very dynamic environments.

Distributed crawling has shown that it can overcome important limitations of the centralized crawling paradigm. However, the distributed nature of current distributed crawlers is currently not fully utilized. The optimal benefits of this approach are usually limited to the sites hosting the crawler. In this work we describe IPMicra, a distributed location aware web crawler that utilizes an IP address hierarchy and allows crawling of links in a near optimal location aware manner. The crawler outperforms earlier distributed crawling approaches without a significant overhead.

Developing the semantic web is a key research challenge. The conference on Ontologies, DataBases, and Applications of Semantics for Large Scale Information Systems (ODBASE’04) provides a forum on ontologies and data semantics that is inclusive of the many computing disciplines involved in such a challenge, such as ontology management, information mining, knowledge representation, information integration, semantic web-services, and text processing. ODBASE 2004 also includes research and interesting descriptions of real-life applications including scale issues in ontology management, information integration, and data mining, as well as papers that examine the information needs of various Web and knowledge applications, including medicine, e-science, history, e-government and manufacturing.

In scientific communication, we usually resort to formal theories – like algebra or first-order logic – to express our thoughts and intuitions in such a way they can be understood by our colleagues. I will argue that the tools of formal ontology (such as the notions of parthood, unity, dependence, identity) can play a similar role in ordinary communication, for instance during e-commerce transactions. I will briefly present what these tools are, and I will give examples concerning their role in facilitating mutual agreement, as well as recognizing and explaining disagreement.

We report on an a set of experiments carried out in the context of the Flemish OntoBasis project. Our purpose is to extract semantic relations from text corpora in an unsupervised way and use the output as preprocessed material for the construction of ontologies from scratch. The experiments are evaluated in a quantitative and ”impressionistic” manner.We have worked on two corpora: a 13M words corpus composed of Medline abstracts related to proteins (SwissProt), and a small legal corpus (EU VAT directive) consisting of 43K words. Using a shallow parser, we select functional relations from the syntactic structure subject-verb-direct-object. Those functional relations correspond to what is a called a ”lexon”. The selection is done using prepositional structures and statistical measures in order to select the most relevant lexons. Therefore, the paper stresses the filtering carried out in order to discard automatically all irrelevant structures.Domain experts have evaluated the precision of the outcomes on the SwissProt corpus. The global precision has been rated 55%, with a precision of 42% for the functional relations or lexons, and a precision of 76% for the prepositional relations. For the VAT corpus, a knowledge engineer has judged that the outcomes are useful to support and can speed up his modelling task. In addition, a quantitative scoring method (coverage and accuracy measures resulting in a 52.38% and 47.12% score respectively) has been applied.

The availability of formal ontologies is crucial for the success of the Semantic Web. Manual construction of ontologies is a difficult and time-consuming task and easily causes a knowledge acquisition bottleneck. Semi-Automatic ontology generation eases that problem. This paper presents a method which allows semi-automatic knowledge extraction from underlying classification schemas such as folder structures or web directories. Explicit as well as implicit semantics contained in the classification schema have to be considered to create a formal ontology. The extraction process is composed of five main steps: Identification of concepts and instances, word sense disambiguation, taxonomy construction, identification of non-taxonomic relations, and ontology population. Finally the process is evaluated by using a prototypical implementation and a set of real world folder structures.

Medicine is one of the best examples of application domains where ontologies have already been deployed at large scale and demonstrated their utility. However, most of the available medical ontologies, though containing huge amounts of valuable information, underlie different design principles as those required by Semantic Web applications and consequently can not be directly integrated and reused. In this paper we describe the generation, maintenance and evolution of a Semantic Web-based ontology in the context of an information system for pathology. The system combines Semantic Web and NLP techniques to support a content-based storage and retrieval of medical reports and digital images.

We address the life cycle of semantic web based knowledge management from ontology modelling to instance generation and reuse. We illustrate through a semantic web based knowledge management approach the potential of applying semantic web technologies in GEODISE, an e-Science pilot project in the domain of Engineering Design Search and Optimization (EDSO). In particular, we show how ontologies and semantically enriched instances are acquired through knowledge acquisition and resource annotation. This is illustrated not only in Protégé with an OWL plug-in, but also in a light weight function annotator customized for resource providers to semantically describe their own resources to be published. In terms of reuse, advice mechanisms, in particular a knowledge advisor based on semantic matching, are designed to consume the semantic information and facilitate service discovery, assembly and configuration in a real problem solving environment. An implementation has demonstrated the integration of the advisor in a text mode domain script editor and a GUI mode workflow composition environment. Our research work shows the potential of using semantic web technology to manage and reuse knowledge in e-Science.

Although ontology has gained wide attention in the area of information systems, a criticism typical of the early days is still rehearsed here and there. Roughly, this criticism says: general ontologies are not suited for real applications. We believe this is the result of a misunderstanding of the role of general ontologies since, we claim, even foundational ontologies (the most general and formal ontologies) have a crucial role in building reusable, adaptable and transparent application systems. We support this view by showing how foundational ontologies can be used in the manufacturing control area. Our approach (partially presented here through an example) provides a domain-specific ontology which is explicitly designed for applications, theoretically organized by a foundational ontology, driven by the application field for all intents and purposes, suitable for communication across different applications.

IPR (Intellectual Property Rights) Management is a complex domain. The IPR field is structured by evolving regulations, practises, business models,...Therefore, DRMS (Digital Rights Management Systems) are very difficult to develop and maintain.The NewMARS DRMS is our contribution to this field. A knowledge oriented approach has been chosen in order to make this development capable of dealing with this complicated domain. This requirement and the objective of easy Web integration have made the Semantic Web technologies the best choice.NewMARS is a semantics enabled metadata managing system. Metadata is associated to IPs (Intellectual Properties) using URIs and it is structured using web ontologies. There are descriptive, rights and e-commerce ontologies for the different views on IPs. Semantic enabled metadata is then used to facilitate content providers to publish intellectual properties offers and customers to find and automatically negotiate purchase conditions.All NewMARS modules are interrelated using the ontologies shared semantics. This has allowed developing very flexible project infrastructures that facilitates easy adaptation to new IP e-commerce scenarios.

Although the first digital archives storing a huge number of resources came into existence years ago, they still lack effective retrieval methods. The most obvious example is the World Wide Web: search engines are improved constantly, however, their hits are still unsatisfactory apart from simplest queries. Most prosperous solutions employ user contexts to estimate the user’s information demand and use this information to deliver more adequate results. In the current paper we introduce the idea of resource context-based information retrieval. In this approach, semantic context description is assigned to each digital resource known to the system and this semantic metadata are exploited by each step during an intelligent search process. Our solution is implemented and evaluated in the VICODI project, as part of a web portal for European history.

The semantics of order-sorted type hierarchies are fundamental both to the retrieval of knowledge from large real-world-size knowledge bases and to the next generation of web technology. The Chrysostom web knowledge-base project provides an interesting case for furthering this research. An attempt to model the social world of the late Roman Empire, this knowledge base rests upon an ontology which uses the knowledge found in a very large body of documents associated with the cities of Antioch in ancient Syria and Constantinople (modern Istanbul) in the fourth and fifth centuries. We describe the knowledge base and its use, as well as the ontology that was created (and continues to develop) to support it.

This paper addresses the issue of simplifying natural language texts in order to ease the task of accessing factual information contained in them. We define the notion of Easy Access Sentence – a unit of text from which the information it contains can be retrieved by a system with modest text-analysis capabilities, able to process single verb sentences with named entities as constituents. We present an algorithm that constructs Easy Access Sentences from the input text, with a small-scale evaluation. Challenges and further research directions are then discussed.

A database federation provides for tight coupling of a collection of heterogeneous legacy databases into a global integrated system. A large problem regarding information quality in database federations concerns achieving and maintaining consistency of the data on the global level of the federation. Integrity constraints are an essential part of any database schema and are aimed at maintaining data consistency in an arbitrary database state. Data inconsistency problems in database federations resulting from the integration of integrity constraints can basically occur in two situations. The first situation pertains to the integration of existing local integrity constraints occurring within component legacy databases into a single global federated schema, whereas the second situation pertains to the introduction of newly-defined additional integrity constraints on the global level of the federation. These situations gives rise to problems in so-called global and local understandability of updates in database federations. We shall describe a semantic framework for specification of federated database schemas based on the UML/OCL data model; UML/OCL will be shown to provide a high-level, coherent, and precise framework in which to specify and analyze integrity constraints in database federations. This paper will tackle the problem of global and local understandability by introducing a new algorithm describing the integration of integrity constraints occurring in local databases. Our algorithm is based on the principle of tight constraining; i.e., integration of local integrity constraints into a single global federated schema takes place without any loss of constraint information. Our algorithm will improve existing algorithms in three aspects: it offers a considerable reduction in complexity; it applies to a larger category of local integrity constraints; and it will result in a global federated schema with a clear maintenance strategy for update operations.

The data integration of a number of local heterogeneous databases, with possible conflicting, mutually inconsistent, information, coming from different places, is an increasingly important issue. In order to avoid such inconsistency, a number of current in-practice developed database systems are based on different software and architectural paradigms, and are specified a number of embedded ad-hoc algorithms for a kind of preferred query-answering w.r.t. some preordering. The query-answering to conjunctive queries is usually performed in two consecutive steps: first are obtained certain answers from the underlying DBMS system, and successively is applied a filtering software, based on particular user-written algorithms, in order to obtain a ‘best subset’ of answers. Thus, the obtained resulting answers does not correspond to the original user’s query: to which kind of logic formula the obtained answers correspond was an open problem. In this paper we show that such bivalent database/software-algorithm paradigm can be unified in an equivalent Abstract Object Type (AOT) database with a partial order, and that the query formula which returns with the same answers, as the answers to a conjunctive query of the original database/softwarealgorithm, is a modal logic formula.

Large-scale database integration requires a significant cost in developing a global schema and finding mappings between the global and local schemas. Developing the global schema requires matching and merging the concepts in the data sources and is a bottleneck in the process. In this paper we propose a strategy for computing the mapping between schemas by performing a composition of the mappings between individual schemas and a reference ontology. Our premise is that many organizations have standard ontologies that, although they may not be suitable as a global schema, are useful in providing standard terminology and naming conventions for concepts and relationships. It is valuable to leverage these existing ontological resources to help automate the construction of a global schema and mappings between schemas. Our system semi-automates the matching between local schemas and a reference ontology then automatically composes the matchings to build mappings between schemas. Using these mappings, we use model management techniques to compute a global schema. A major advantage of this approach is that human intervention in validating matchings mostly occurs during the matching between schema and ontology. A problem is that matching schemas to ontologies is challenging because the ontology may only contain a subset of the concepts in the schema or may be more general than the schema. Further, the more complicated ontological graph structure limits the effectiveness of some matchers. Our contribution is showing how schema-to-ontology matchings can be used to compose mappings between schemas with high accuracy by adapting the COMA schema matching system to work with ontologies.

In this paper a stepwise methodology for ontology alignment and merging is proposed. The knowledge model underlying this methodology is based on an extended version of the initial Dogma framework that is a database inspired approach to ontology engineering. The methodology we propose in this paper encompasses several techniques and algorithms that can be used and combined in order to assist the user with the integration of ontologies. We explain how some of these algorithms make use of already existing thesauri in order to provide the user with useful suggestions. The implementation of these algorithms has resulted in a tool that clearly visualises the overall ontology integration process.