Abstract: A system and method for determining a quantitative measure of search efficiency of related Web pages. An information goal is specified. A target Web page is identified within a plurality of Web pages. The information goal is searched via a search function in the Web pages to identify potential Web pages that include at least one hyperlink referencing and proximal cues relating to distal content included in another potential Web page. An activation network is formed. A directed graph is built, including nodes corresponding to the potential Web pages and arcs corresponding to the hyperlinks. A weight is assigned to each arc to represent a probability of traversal of the corresponding hyperlink based on a relatedness of keywords in the information goal to the proximal cues. A traversal through the activation network to the node corresponding to the target Web page is evaluated as a quantitative measure of search efficiency.

Abstract: A method and system for autonomic relevancy building using information obtained from clients to improve customer support. Upon receiving a client request for product support information, product support information objects representing the requested product support information in an information library are identified. Relevant product support information corresponding to the identified relevant product support information objects are provided to the client. Data regarding whether or not the information provided to the client is relevant may be obtained from the client through implicit feedback (such as, system usage), explicit feedback (such as, user feedback), metadata information, or any combination thereof. Relevancy associations may be created between the product support information objects based on the relevancy data. Relevancy weights may also be assigned to each relevancy association based on the relevancy data, wherein the relevancy weights designate a certainty of each relevancy association.