REWERSE I5 - Evolution and Reactivity
Mission
To define declarative languages, tools and methodologies for specifying and dealing with reactivity, evolution and propagation of changes in the Web.
In this page
About WG I5
Besides the realization (i.e. technologies and languages) and the use (i.e. querying) of the Semantic Web, its maintenance and evolution are important issues. The Semantic Web is a "living organism", which combines autonomously evolving data sources/knowledge repositories. This dynamic character of the Semantic Web requires (declarative) languages and mechanism for specifying its maintenance and evolution. For example, for changing the behaviour of a data source so that a new rule becomes in effect, one should not have to be concerned with the complex, inter-related, and dynamically obtained knowledge in the various data sources, and have a way to simply specify how the knowledge is to be changed. This not only requires the existence of a language for specifying such changes (or updates), but also mechanisms for consistently handling and propagating the updates.
Semantic Web data sources may be reactive, not only due to the incorporation of updates, but also in that they perceive events and incoming messages, communicate with other components, and execute actions. For example, an update with a customer's new address could impose a reactive behaviour in numerous databases that need to be changed. Another example of reactivity, in the area of bioinformatics, is an addition of new sequences in a sequence database, which needs to be fed into other derived databases. This mechanism is also important at the meta-level, where updated data formats and schemata need to be applied to dependent data sources. In general, reactive behaviour is useful to mediate between data sources and to implement workflows integrating various data sources and spreading information through a network of data sources.
In this working group of REWERSE we investigated on all these aspect of evolution and reactivity in the (Semantic) Web.
Brief Description of Work
In the Semantic Web knowledge is stored in various autonomous data sources, which evolve in time. Thus, besides the realization and the use of the Semantic Web, its maintenance and evolution are important issues. In response to updates in data sources, these may have to react and propagate the changes. Besides this form of reaction, Semantic Web data sources may have to react in other aspects, e.g.: by perceiving events and incoming messages; by communicating to other components; by executing actions.
The WG I5 of REWERSE will focus on these aspects of evolution and reactivity of the Semantic Web, and pursue the following main goals:
- investigate the evolution of data, rules, and ontologies on the
web concerning:
- updates, changes, and revisions
- dealing with potentially resulting inconsistencies (in close collaboration with WG I4)
- the relation of local and global views in the light of evolution (in collaboration with WG I4)
- investigate reactivity on the web in particular concerning:
- specification of reactive behaviour with reaction rules
- execution and verification of reaction rules
- communication infrastructure and communication models
- develop case studies scenarios for both evolution and reactivity behaviour
- build prototypes for both showing how to manage and specify evolution, and to cater for reactivity on the Semantic Web.
The WG I5 Flyer can be found here.
WG I5 Deliverables
The following WG I5 Deliverables were produced during the REWERSE project:
- Deliverables due at month 6
# I5-D1 Title State-of-the-art on evolution and reactivity Description A report comprising the current state-of-the-art on evolution and reactivity on the semantic web will be written. This comprises a summary of the literature on event-condition-action rules, active databases, agent-based systems, agents and logic programming, updates and belief revision, communication means in the semantic web, dependency analysis, rule mark-up standards. - Deliverables due at month 12
# I5-D2 Title Use-cases on evolution Description Development of a case study scenario and a representative collection of use cases for communication and evolution of data sources in the semantic web. The use cases will be developed where possible in collaborations with the applications' working groups. # I5-D3 Title Use-cases on reactivity Description Development of a case study scenario and a representative collection of use cases for reactive behaviour on the semantic web. The use cases will be developed where possible in collaborations with the working group on bioinformatics and personalized information systems. - Deliverables due at month 18
# I5-D4 Title Specification of a model language and architecture for reactivity and evolution Description Based on a model for local evolution and reactivity (related to those for active data-bases and agents), a model for propagation of changes, reactivity, and global evolution model in the Web and the Semantic Web will be specified. The specification will pro-vide a global architecture, description of communication issues, and a set of modular languages. The languages will be in the form of event-condition-action rules, defined in consultation with working group I1. In particular, the syntax needs to reflect various types of actions (local and remote updates, general (black-box) messages, a notion of events, and event languages, and conditions (that are expressed as queries). - Deliverables due at month 30
# I5-D5 Title A first prototype on evolution and behaviour at the XML level Description A first prototype of the language and architecture defined on the previous deliverable will be implemented, and illustrated in a simple demonstrator application chosen among those described in the use cases. - Deliverables due at month 36
# I5-D6 Title Reactive rule ontology: RDF/OWL level Description Definition of a concrete ontology for reactive rules at the RDF/OWL level, in line with the general framework defined in deliverable I5-D4, and gaining from possible feedback and requirements from other workgroups. # I5-D7 Title Completion of the prototype scenario Description Refinement and tests of the prototypes, and their usage in use case scenarios. - Deliverables due at month 42
# I5-D8 Title Realistic information integration application Description Development of realistic information integration application in bioinformatics in collaboration with working group A2. # I5-D9 Title Prototype on the RDF/OWL level Description Extension of the prototypes to deal with evolution and reactivity at the RDF/OWL level, and in accordance with the ontology previously defined. - Deliverables due at month 48
# I5-D10 Title Pre-standardization of the language Description Pre-standardization of language, semantics, processing model.
WG I5 Publications
To find any of the WG I5 related publications please consult the general list of REWERSE publications and filter them by Workgroup I5. The web page of all REWERSE publications is the following:
WG I5 Sub-projects and Prototypes
- MARS -
Modular Active Rules for the Semantic Web
MARS is a general framework for specifying and implementing active rules (ECA rules) in the Semantic Web. SWAN (Semantic Web Active Node) is an architecture for individual active nodes that are compatible with MARS. - r3 - Resourceful Reactive Rules
The r3 framework is capable of dealing with reactive rules that use different languages at the rule component level. This language heterogeneity at the top level is recursivelly extensive to each of the different components (event, condition or action). - REX -
Rule and Event eXplorer
REX is designed as a front end to the timed automata CASE tool Uppaal allowing users to specify rules and events in REX instead of timed automata models inUppaal but still take advantage of the power of model-checking. - XChange - Rule language for reactive behavior on the
Web
Following a declarative approach to reactivity on the Web, a novel reactive, rule-based language called XChange is being developed, covering: updating data on the Web, exchanging information about events (such as executed updates) between Web nodes, and reacting to combinations of such events.
REWERSE Institutions that participated in WG I5
WG I5 Coordination Team
- Coordinator: José Alferes, Lisbon.
- Deputy coordinator: Wolfgang May, Göttingen.
- Assistant: Ricardo Amador, Lisbon.
