Corporate Memory Management through Agents
The main objective of the project is to implement and trial a Corporate Memory Management framework based on Agent technology.
Knowledge Management is becoming increasingly important for a correct, efficient and effective exploitation of the Know How available to an enterprise.
The project will assess two scenarios with real end users:
insertion of new employees in the company,
detection of technology movements and diffusion among employees.
The main innovative aspect of the project will stem from he seamless integration of several technologies: a multi-agent system, Enterprise models and User models, based on ontology, machine learning techniques, exploitation of XML standards.
The main objective of the project is to implement and trial a corporate memory management framework based on agent technology, and will address particularly the following issues:
Enhancing the insertion of new employees in the company,
Performing process that detect, identify and interpret technology movements and interactions for matching technology evolutions with market opportunities to diffuse among employees innovative ideas related to technology monitoring activities.
Moreover, the project will give opportunities to merge emergent technologies like Multi-Agent System, XML and RDF formats, machine learning techniques and to apply them to corporate memory concept.
Each WorkPackage will provide a well defined part of the work.
WP1 will ensure the technical coordination of the project activities. It will be also responsible of external relations with other projects and dissemination of all results.
WP2 will provide enterprise and users models.
WP3 will design and implement generic agent dedicated to the information retrieval by taking into account requirement addressed by WP 4 and WP 5. This work will concern the definition of a multi-agent architecture and the specification of the communication and transaction protocols.
WP4 will study techniques for XML document classification and retrieval by exploiting ontologies, enterprise and user models delivered by WP2. It will also develop inference mechanisms on ontologies, to improve the effectiveness of these agents.
WP5 will focus on machine learning field. Its main aim will be to give adaptive capabilities to the agents that will be in touch with the user.
WP6 will gather all the software components provided by other WP and will achieve assessment of the whole project by implementing two prototypes:
An intermediate prototype to have first results and to provide feedback to other workpackages in order to improve models and techniques.
A final prototype to be used as a demonstrator in order to show improvements brought by the project on concerned areas.
The assessment phase will rely on two particular scenarios:
The first one concerns the learning and the integration of a new employee,
The second one the retrieval of innovative data.