General purpose centralized whole Web indexes have become increasingly harder to set up and maintain with the scale problem of Web size. In this article we present a novel approach to information indexing and searching based on partitioning the Web according to Digital Neighbourhoods (DNs). Four concepts of DN are introduced which allow the definition of indexing and searching scopes. The architecture of a DN-based distributed system is then presented, and its characteristics of effective indexing, scalability and interoperability are discussed. The structure of a working prototype and some implementation details are presented. 1. INTRODUCTION With the growth of the World-Wide Web (Web), the task of locating and retrieving relevant information is facing scale problems that tend to increase [Berghel, 1996]. Related problems exist in the corporate intranet arena [Strom, 1995], in Digital Libraries [Paepcke et al., 1996], and in information gathering digital assistants [Balabanovic and S...

The evaluation of an implication by Imaging is a logical technique developed in the framework of modal logic. Its interpretation in the context of a "possible worlds" semantics is very appealing for IR. In 1994, Crestani and Van Rijsbergen proposed an interpretation of Imaging in the context of IR based on the assumption that "a term is a possible world". This approach enables the exploitation of term-term relationships which are estimated using an information theoretic measure. Recent analysis of the probability kinematics of Logical Imaging in IR have suggested that this technique has some interesting sense resolution properties. In this paper we will present this new line of research and we will relate it to more classical research into word senses.

The task of indexing and searching information in the World-Wide Web face scale problems with Web size and growth. We argue that this is due, in part, to the centralized nature of Web indexing systems currently in use. This paper proposes an open distributed Web View-based system to deal with these problems. The concept of Web View is introduced, the system architecture is presented, and it is shown how the system is able to effectively build, store and search distributed indices. The components of the system modules are then described, and a working prototype is presented. 1. INTRODUCTION Two basic strategies are currently used to build World-Wide Web (Web) indices: manually updated resource lists (e.g., [Yahoo!]), and mechanically updated index bases (e.g., [Altavista] and [Excite]). The available implementations of both strategies are essentially centralized as their processes of collecting information, and storing and searching indices are performed into/from a single database or ...

The evaluation of an implication by Imaging is a logical technique developed in the framework of modal logic. Its interpretation in the context of a “possible worlds ” semantics is very appealing for IR. In 1994, Crestani and Van Rijsbergen proposed an interpretation of Imaging in the context of IR based on the assumption that “a term is a possible world”. This approach enables the exploitation of term– term relationships which are estimated using an information theoretic measure. Recent analysis of the probability kinematics of Logical Imaging in IR have suggested that this technique has some interesting sense resolution properties. In this paper we will present this new line of research and we will relate it to more classical research into word senses. 1