Efficient Interactive Fuzzy Keyword Search

Presentation

Microsoft PowerPoint (646Kb)

Published Version

Preview

PDF (1278Kb)

Abstract

Traditional information systems return answers after a user submits a complete query. Users often feel “left in the dark” when they have limited knowledge about the underlying data, and have to use a try-and-see approach for finding information. A recent trend of supporting autocomplete in these systems is a first step towards solving this problem. In this paper, we study a new information-access paradigm, called “interactive, fuzzy search,” in which the system searches the underlying data “on the fly” as the user types in query keywords. It extends autocomplete interfaces by (1) allowing keywords to appear in multiple attributes (in an arbitrary order) of the underlying data; and (2) finding relevant records that have keywords matching query keywords approximately. This framework allows users to explore data as they type, even in the presence of minor errors. We study research challenges in this framework for large amounts of data. Since each keystroke of the user could invoke a query on the backend, we need efficient algorithms to process each query within milliseconds. We develop various incrementalsearch algorithms using previously computed and cached results in order to achieve an interactive speed. We have deployed several real prototypes using these techniques. One of them has been deployed to support interactive search on the UC Irvine people directory, which has been used regularly and well received by users due to its friendly interface and high efficiency. answers. This information-access paradigm requires the user to have certain knowledge about the structure and content of the underlying data repository. In the case where the user has limited knowledge about the data, often the user feels “left in the dark” when issuing queries, and has to use a tryand-see approach for finding information, as illustrated by the following example. At a conference venue, an attendee named John met a person from a university. After the conference he wanted to get more information about this person, such as his research projects. All John knows about the person is that he is a professor from that university, and he only remembers the name roughly. In order to search for this person, John goes to the directory page of the university. Figure 1 shows such an interface. John needs to fill in the form by providing information for multiple attributes, such as name, phone, department, and title. Given his limited information about the person, especially since he does not know the exact spelling of the person’s name, John needs to try a few possible keywords, go through the returned results, modify the keywords, and reissue a new query. He needs to repeat this step multiple times to find the person, if lucky enough. This search interface is neither efficient nor user friendly.