Generating associative ripples of relevant information from a variety of data streams by throwing a heuristic stone

Abstract

Recently, the vast dialog in the microblog, such as twitter, Facebook has become increasingly popular. As we post more messages in microblogs, information is spreading more quickly and widely. These widely spread and diversified contents could be viewed as data streams, which have become an important part of the Internet resources. However, these separated data streams are littery and meaningless, so we need to collect and organize them together to provide us with meaningful information. It is hard to imagine that we could find useful information by simply inputting a few keywords into a search engine in such a stream environment. In this study, we try to find a way to seek the information related to users' personal and current interests and needs among these data streams and provide users with other more relevant information. We introduce a set of metaphors to represent a variety of data streams in different levels, and define two new metaphors: heuristic stone and associative ripple to assist the seeking process and describe the results. Based on these, we further propose two algorithms for the information seeking and processing, and discuss a scenario of the information seeking process that utilizes the proposed metaphors and algorithms.