ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Useful Query Facets Extracting Automatically from Top Retrieved Documents by Using QDMiner System

Journal: IPASJ International Journal of Information Technology (IIJIT) (Vol.5, No. 10)

Publication Date:

Authors : ;

Page : 001-004

Keywords : ;

Source : Downloadexternal Find it from : Google Scholarexternal

Abstract

ABSTRACT Web search for queries are often ambiguous or multi-faceted, which makes a simple ranked listing of results insufficient. To aid records finding for such faceted queries, we explore a method that explicitly represents exciting aspects of a query the use of agencies of semantically associated phrases extracted from search results. As an example, to the query “baggage allowance”, these businesses might be exclusive airways, exceptional flight nature (home, global), or specific journey classes (first, commercial enterprise, financial system); we call those groups query facets and the phrases in these agencies' aspect phrases. We expand a supervised method primarily based on a graphical version to apprehend query facets from the noisy candidates determined. The graphical version learns how possibly a candidate time period is to be a facet term as well as how probable two terms are to be grouped together in a query side, and captures the dependencies between the 2 elements. We propose a systematic solution, which we check with as QDMiner, to automatically mine query aspects by way of extracting and grouping common lists from loose textual content, HTML tags, and repeat areas inside top search effects. Compared to earlier works on constructing facet hierarchies, our method is unique in factors: Open domain and Query dependent. Keywords: QDminer, ODP Taxonomy, Query Facets, Open Domain

Last modified: 2017-11-12 23:56:23