P I S A: A Personalized Information Search Assistant

Maria Elena Renda
Ph.D. Thesis
Dept. of Information Engineering
Scuola Superiore Sant'Anna di Studi Universitari e di Perfezionamento
Pisa - Italy

Istituto di Informatica e Telematica (IIT)
Consiglio Nazionale delle Ricerche (CNR)
I-56100
Pisa (PI) ITALY

Contact:
Elena.Renda_AT_iit.cnr.it

Abstract. A common characteristic of most of the traditional search and retrieval systems is that they are oriented towards a generic user, often failing in connecting people with what they are really looking for.
Suppose to search the Web for the acronym “IR”; the results returned include investor relations web pages, pages on the iridium element, information retrieval pages, pages on infrared radiation and light, and so on. If the same query is submitted by different users to a typical search engine, it will probably return the same result, regardless of who submitted the query. This is because search tools are designed to satisfy people in general, not the searcher in particular.
Personalization can be defined as an approach aimed at tailoring the information and services to match the unique and specific needs of an individual user, thus improving the retrieval process.

In this thesis we propose a model for personalizing a search system in order to address this need, so that, for instance, the businessman searching for “IR” will find at the top of the results investor relations web pages, the chemist will find information related to the iridium element or the infrared radiation, the information retrieval researcher will find one of the “Information Retrieval” books published.
In particular, we have modeled and designed P I S A, a Personalized Information Search Assistant which, rather than relying on the unrealistic assumption that the user will precisely specify what she is really looking for when searching, leverages implicit information about the user's interests. P I S A provides the user with a highly personalized information space where she can create, manage and organize folders (like, for instance, in email programs), and manage documents retrieved by the system into her folders to best fit her needs. Furthermore, P I S A offers different mechanisms to search the Web, and the possibility of personalizing the result delivery and visualization.
P I S A learns user and folder profiles from user's choices, and these profiles are then used to improve retrieval effectiveness in searching, by selecting the relevant resources to query and filtering the results accordingly. A working prototype has been also developed, tested and evaluated.

The Personalized Information Search Assistant proposed in this thesis is a desktop application which provides personalized information organization, search, presentation and delivering features, helping the user in finding relevant information among the information resources. We believe that from such a personalized system users could benefit for finding relevant information for their interests in a broad sense, gaining in time, quality of the documents and information retrieved, and satisfied information needs.