Thesis project: Services and User Information Retrieval Agents Integration in a

Thesis project: Services and User Information Retrieval Agents Integration in a Digital Library.

Synopsis

This thesis is focused in the topics of digital libraries and user agents, proposing a solution based on these in order to solve the problem of services integration and information retrieval in a digital library. The subject of digital libraries is at the present time an important research area where numerous works have been writed; the main characteristics of these libraries are the fact of being a vast, dynamic and complex deposit of information, that is well organized and structured, and that includes as part of their infrastructure the services and tools necessary for the exploitation and manipulation of this information. The specific case of a floristic digital library with all these characteristics makes an excellent test area to apply and develop an architecture that supports the integration of services. Among the group of services included in the digital library, user agents appear to be a good solution in order to carry out all type of activities on the information of the library due to their capacity to work in autonomous or semi-autonomous ways to accomplish specific tasks on behalf of their users.

Among the numerous problems that could appear in the handling and structuring of a digital library are those related to the integration of their services, this is why it arises the idea of proposing and developing an architecture that solves efficiently this problem. Mobots is an architecture for a floristic digital library designed to integrate and support diverse services and agents through a common interface. The Mobots architecture is mainly focused in the control of agents that are created like instances of classes of agents that have been already registered in the system. These instances could also be manipulated through Mobots in order to change their execution state.

As part of the group of agents that is supposed to conform the services of the digital library, the Flora Retrieval Agent (FRA) was developed for the purpose of information retrieval. FRA makes use of the "knowledge" it has on the user interests in order to carry out searches and recovery of information in semi-autonomous ways -this "knowledge" is stored in user profiles. Another of their important characteristics is the notification to the user of new information in the moment it arrives to the library, avoiding with this invested and lost hours of the user in long searches.