While current search engines serve known-item search such as homepage finding very well, they generally cannot support exploratory search or information foraging effectively. In exploratory search, users do not know their information needs precisely and also often lack the needed knowledge to formulate effective queries, thus querying alone, as supported by the current search engines, is insufficient, and browsing into related information would be very useful. Currently, browsing is mostly done through following hyperlinks embedded on Web pages navigating through structures consisting of a fixed set of categories or other meta-data available, which tends to be restrictive.

In this project, we propose to leverage search logs to allow a user to go beyond hyperlink browsing and support flexible browsing through a multi-resolution topic map (see Fig. 2) constructed based on search logs. A topic map is an analogy of geographical map, such as a map in a park shown on the right part (see Fig. 1), in an information space. Such a topic map can support a user to zoom in, zoom out, and navigate horizontally over the information space, and thus provide flexible and effective browsing capabilities for end users.

Specifically, we treat search logs as "footprints" left by previous users in the information space and build a multi-resolution topic map to semantically capture and organize them in multiple granularities.

Fig. 1: Geographical Map

An online demo is available at here (see Fig. 3 for the interface). In our system, we support:

• Querying. A user can submit a keyword query using the querying box on the top. The system would then fetch the results from a commercial search engine and show them on the right pane. Furthermore, we treat the current query as a map node in the left pane and locate its surrounding topic map nodes: parents, children, and horizontal neighbors.
• Browsing a map node. A user can click a map node in the left pane if it is appealing to the user. After the node is clicked, the left pane will be updated and a local map surrounding the clicked map node will be displayed. These nodes provide ways for users to zoom out to a parent node, zoom in to a child node, or move horizontally to a neighbor node. At the same time, the topic region corresponding to the current map node is displayed on the right pane. Currently, our topic regions are defined as the click-throughs and search results of the corresponding query of the clicked node. In a topic region, both click-throughs and search results are ranked.
• Browsing a hyperlink. As usual, when a user finds that a hyperlink is appealing (based on the user's perceiving of urls or snippets), the user can click on the hyperlink and the right pane will be refreshed to display the page content.

The unique feature in our system is the topic map. It can support effective browsing for end users when surfing the Web. As new users use the map for navigation, they leave more footprints, which can then be used to enrich and refine the map dynamically and continuously for the benefit of future users. Thus, by turning search logs into a topic map, we can establish a sustainable infrastructure to facilitate users to surf the information space in a social and collaborative manner.