DocumentCode :
3585916
Title :
Ranking model adaptation for domain specific mining using binary classifier for sponsored ads
Author :
Krishnamurthy, M. ; Pillai, Anitha S. ; Anuja Jaishree, N.A. ; Kannan, A.
Author_Institution :
Dept. of CSE, KCG Coll. of Technol., Chennai, India
fYear :
2014
Firstpage :
35
Lastpage :
42
Abstract :
Domain - specific search focuses on one area of knowledge. Applying broad based ranking algorithms to vertical search domains is not desirable. The broad based ranking model builds upon the data from multiple domains existing on the web. Vertical search engines attempt to use a focused crawler that index only relevant web pages to a predefined topic. With Ranking Adaptation Model, one can adapt an existing ranking model of a unique new domain. The binary classifiers classify the members of a given set of objects into two groups on the basis of whether they have some property or not. If it is property of relevancy, it is returned to the search query of that particular domain vertical. Sponsored ads are then placed alongside the organic search results and they are ranked with the help of bid, budget and quality score. The ad with the highest bid is placed first in the ad listings. Later, the ad with a maximum quality score is found by click through logs which is replaced in first position. Thus, both organic search and sponsored ads are returned for the specific domain, making it easy for the users to get access to real time ads and connect directly with advertisers as well as to get information on the search query.
Keywords :
Internet; data mining; pattern classification; query processing; search engines; support vector machines; SVM; Web page; binary classifier; broad based ranking model; domain specific mining; ranking adaptation model; search engine; search query; sponsored ads; support vector machine; Adaptation models; Classification algorithms; Data models; Databases; Search engines; Support vector machines; Web pages; Information Retrieval; Keyword Auction; Learning to Rank; Online Algorithm; Pay per click; Sponsored Ads;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Hybrid Intelligent Systems (HIS), 2014 14th International Conference on
Print_ISBN :
978-1-4799-7632-4
Type :
conf
DOI :
10.1109/HIS.2014.7086171
Filename :
7086171
Link To Document :
بازگشت