Biologically Motivated Distributed Designs for Adaptive Knowledge Management

TL;DR

This paper presents biologically inspired distributed systems, TalkMine and @ApWeb, that adaptively improve information retrieval by learning user categories and structural relationships in data, enabling active, evolving user interactions.

Contribution

It introduces two novel adaptive recommendation systems based on biological network metaphors, enhancing knowledge management in distributed information systems.

Findings

TalkMine learns and adapts keywords based on user communities.

@ApWeb adjusts citation and hyperlink structures to user expectations.

The combined system enables active, evolving interactions with information resources.

Abstract

We discuss how distributed designs that draw from biological network metaphors can largely improve the current state of information retrieval and knowledge management of distributed information systems. In particular, two adaptive recommendation systems named TalkMine and @ApWeb are discussed in more detail. TalkMine operates at the semantic level of keywords. It leads different databases to learn new and adapt existing keywords to the categories recognized by its communities of users using distributed algorithms. @ApWeb operates at the structural level of information resources, namely citation or hyperlink structure. It relies on collective behavior to adapt such structure to the expectations of users. TalkMine and @ApWeb are currently being implemented for the research library of the Los Alamos National Laboratory under the Active Recommendation Project. Together they define a biologically motivated information retrieval system, recommending simultaneously at the level of user knowledge categories expressed in keywords, and at the level of individual documents and their associations to other documents. Rather than passive information retrieval, with this system, users obtain an active, evolving interaction with information resources.