Paper Info
Title
Ranking web pages by associating keywords with locations
Abstract
Many Web queries contain both textual keywords and location words. When answering such queries, the association between the textual keywords and locations in a Web page should be taken into account. In this paper, we present a new ranking algorithm for location-related Web search, which is called MapRank. Its main idea is to extract the associations between keywords and locations in Web pages and further use them to improve ranking effectiveness. We first determine map each keyword with specific locations and form a set of pairs. Then, we compute the location-constrained score for each keyword and combine it into the ranking procedure. We conduct comparison experiments on a real dataset and use the metrics including MAP and NDCG to measure the performance of MapRank. The results show that MapRank is superior to previous methods with respect to different symbolic-location-related queries.
Year
DOI
Venue
2013
10.1007/978-3-642-38562-9_62
WAIM
Keywords
Field
DocType
ranking effectiveness,new ranking algorithm,location-related web search,textual keyword,comparison experiment,web query,ranking web page,ranking procedure,web page,location word,different symbolic-location-related query
Data mining,Learning to rank,Ranking,Ranking SVM,Information retrieval,Web page,Computer science
Conference
Citations 
PageRank 
References 
0
0.34
14
Authors
5
Name
Order
Citations
PageRank
Peiquan Jin133854.93
Xiaoxiang Zhang231.47
Qingqing Zhang310214.76
Sheng Lin491.92
Lihua Yue534046.44