RICERCA ITALIANA     

Research program

ESTEEM: Emergent Semantics and cooperaTion in multi-knowledgE EnvironMents - Advanced methods and tools for semantic cooperation in Web virtual communities

University Co-ordinator

Università degli Studi di BRESCIA - ELETTRONICA PER L'AUTOMAZIONE - BRESCIA(BS)

Research Unit Leader

Valeria DE ANTONELLIS

Description

Main goal of the research activity of the team at University of Brescia (UNIBS) is the development of advanced methods, techniques and tools to support semantic service discovery in collaborative multi-knowledge environments. A possible application scenario is given involving medical personnel across several countries that need to share electronic medical information in order to enhance decision-making processes and decrease medical errors. The capability of accessing available distributed services using shared ontologies provides a great advantage for the integration of disparate hospital information systems, as well as the possibility of providing more accurate diagnoses and a well organised knowledge base for sharing, tutoring and researching. As an example, a doctor could input a patient's illness and obtain all relevant information about it, from the genes that may contribute to causing it, to the symptoms and possible treatments. Proper discovery tools could identify suitable available services faciltating medical staff to investigate, annotate, and analyse the data (e.g. services for viewing and annotation of various types of images, for searching of patient data, for image analysis and data analysis).
Effective service discovery methods and techniques under highly dynamic and context-dependent requirements are primary needs in collaborative multi-knowledge environments. The emergence of semantics is a key issue to enforce timely discovery and dynamic composition of distributed services. Based on Web Service technologies and available standards (like WSDL for service description, UDDI for service registry, SOAP for message exchange and BPEL4WS for service orchestration), Semantic Web and, more recently, Semantic Grid tools are being developed. In the future, it is envisaged that application development will be mostly based on the composition of services published and made available in collaborative multi-knowledge environments. Modern approaches for service discovery have to address the treatment of dynamical aspects both with respect to the continuous addition and removal of services in a highly variable environment and with respect to different contexts in which a service could be invoked.
Advanced techniques and tools for enabling dynamic semantic service discovery are therefore highly desired and required.

In the ESTEEM project, UNIBS will develop a unified service oriented framework to support semantic service discovery and use in a timely and flexible fashion. The framework will be based on: i) service semantic description techniques, based on domain and service ontologies; ii) ontology-based matching techniques based on deductive reasoning and similarity, for service selection on the basis functional and non functional requirements; iii)flexible composition techniques taking into account dynamic aspects of service evolution.
To support the defined framework, UNIBS will develop corresponding prototype tools to be deployed as services on the reference P2P middleware architecture defined in the project. UNIBS will develop in the ESTEEM demonstrator modules concerning service discovery issues.

The workplan includes two phases. Each phase consists of activities and related deliverables. Phases are described in the following.

Phase 1

Activity 1.1

The first activity will regard a comparative study and analysis of existing service discovery approaches with the aim of highlighting weaknesses and strengths with respect to the ESTEEM project. The current approaches to service discovery will be surveyed in particular for analysing the crucial aspects of emergent semantics, i.e. service semantic description, discovery and use, in collaborative multi-knowledge environments. Particular focus will be on service semantic matching considering both functional and non functional properties of services, ontology-based deductive reasoning and similarity, dynamic aspects of service evolution. All the above issues will be analysed considering specific requirements of collaborative multi-knowledge environments in open networked contexts, like semantic Grids and peer-based systems, characterized by a set of independent peer parties without prior reciprocal knowledge and no degree of relationship, that dynamically need to cooperate by sharing data and services.
In particular, the following issues will be carefully identified and explored:
- Timeliness, that is, services need to be composed as and when they are required. It is important that automated tools are available to support or even enact this endeavour. Thus research is required to develop general techniques and tools that enable the ability to discover semantically appropriate services, the ability to select the best set of offers at a given moment.
- Context-adaptation, that is, services need to be presented to users at the right time, in the right combination, and with the right level of quality.
- Autonomic behaviour, that is, services should auto-configure to meet the needs of their multiple users in dynamically changing circumstances, and self-heal in the presence of faults. The system should also support evolutionary growth as new services become available.

- Results of activity 1.1

D3.1: Technical report illustrating the state of the art on the existing methods and techniques for (i) service discovery in networked systems, semantic Grids and P2P systems, (ii) ontology-based matching based on deductive reasoning and similarity(iii) service evolution and autonomic behavior.

Activity 1.2

The second activity will be devoted to the collection of detailed requirements of the ESTEEM application scenario (scientific
collaboration in medicine).
Collected requirements will be exploited for the integrated demonstrator that will be implemented in phase 2. We point out that, even if prototype demonstration is on a specific application scenario, the ESTEEM platform is general purpose and can support different semantic communities.

- Results of activity 1.2

DALL.1 (jointly developed by all partners): Technical report illustrating all the requirements collected for the application scenario.

Activity 1.3

The third activity will be devoted to the design of advanced methods and techniques for the semantic service discovery and use in collaborative multi-knowledge environments for semantic collaboration and interoperation.
UNIBS research activity will be devoted to the development of: ontology-based matching techniques based on deductive reasoning and similarity; flexible composition techniques taking into account dynamic aspects of service evolution. Furthermore, service semantic description techniques, based on domain and service ontologies, previously developed by UNIBS for semantic service description and semantic service comparison will be used and adapted to this context. They will be the starting point for developing advanced techniques addressing timeliness, context-adaptation and autonomic behavior issues.
Such advanced techniques will be ontology-based. The use of ontologies specifically enables service matchmaking in the discovery process. In fact, the elements used for service capability description refer to concepts that can be properly defined and semantically related in domain ontologies. Semantic relationships between concepts can be exploited to establish the type of matching between advertisements and requests. For the service discovery process we intend to follow a hybrid multimode matching approach, combining different matching techniques (e.g., a deductive capability matching extended with a flexible similarity evaluation scheme). The resulting approach should provide service advice at multiple levels of granularity with rating of adviced services according to different kinds of comparison strategies.
As for flexible service composition techniques, methods for representing dynamics of service evolution will be investigated. In fact, in collaborative multi-knowledge environments services continually appear and disappear according to changes in the community composition. Service substitutability must be supported and flexibility in assembling and coordinating services has to be enforced.

- Results of activity 1.3

D3.2: Technical report defining a formal framework for semantic service matching in collaborative multi-knowledge environments, provided with ontology-based matching techniques based on deductive reasoning and similarity, for service discovery considering both functional and non functional aspects, and flexible composition techniques taking into account dynamic aspects of service evolution.

Activity 1.4

This activity will concern the high-level specification of the ESTEEM architecture, to be defined in cooperation with the other partners of the project. The architecture will provide a set of services for integrated and trust-aware access to data distributed over peers of a semantic community, a set of services for the discovery and matching of web services within the community, and a set of services for consensus-driven formation and management of semantic communities. In particular, the UNIBS activity will be devoted to the definition of the WSDL interfaces of the set of services for the discovery and matching of web services within the community.

- Results of activity 1.4

DALL.2(jointly developed by all partners): Technical report defining the high-level architecture of ESTEEM, and the WSDL specification of the provided services.

Phase 2

Activity 2.1

Goal of UNIBS research activity is the design of a detailed architecture of support services for service discovery in collaborative multi-knowledge environments, using domain and service ontologies as knowledge spaces for semantically describing shared resources, and ontology-based matching techniques for service semantic interoperability. The design activity will involve the development of a mock-up for the services to be developed by UNIBS for the discovery and matching of web services within the community. UNIBS will be involved with other partners in a common testing process, specifically for the aspects related to own prototypes. The results of this test shall be fully taken into account in the next activity 2.2.

- Results of Activity 2.1

DALL.3: (jointly developed by all partners) Technical report defining the detailed architecture of ESTEEM.
DALL.4: (jointly developed by all partners) Mock-up prototypes and data collected during the testing phase.

Activity 2.2

In this activity the implementation of the UNIBS services will take place based on the specification and design developed in the previous activities. The implementation approach will be based on web services and grid technologies.

- Results of Activity 2.2
D3.3: Implementation of a set of services for services for the discovery and matching of web services within the community.

Activity 2.3

This activity will be jointly developed by all partners. It consists the implementation of an integrated demonstrator, showing the usage and feasibility of the ESTEEM platform in a real case.
The implementation of the ESTEEM demonstrator will be based on the services developed in Activity 2.2. UNIBS will be in charge of developing those modules of the demonstrator involving service discovery issues.

- Results of Activity 2.3

DALL.5(jointly developed by all partners): Technical report illustrating the integrated demonstrator, as well as the results of the experimentation in the context of the application scenario.