Query Processing Algorithms In Searching Advertising

Searching advertising has evolved as the delivery of relevant, targeted text ad-vertisements for Web queries. To match the most relevant advertisements for queries,query processing algorithms were deeply researched during previous works, such asbool retrieval and query expansion.While most of current state-of-the-art algorithms appeal to Web search results asexternal resources to expand queries, we propose a novel approach based on Wikipediafor query augmentation against rare queries in search advertising. By retrieving thetop- relevant articles in Wikipedia with Web query, we can extract more representa-tive information and form a new ad query for the web query. With the new ad query,more relevant advertisements can be identi?ed. We design a set of experiments and theresults turn out that our approach is very e?ective for rare queries.Because WAND can exhibit a good control over the result number, we replacethe tranditional bool relation with it. In this paper, we develop a novel frameworkthat can substantially enhance the e?ciency of WAND processing. This frame- workfocuses on two attributes in WAND: term weight and WAND threshold. To begin with,we propose the notion of mandatory terms indicating those query terms with very highweights, andpresentasearch-basedmethodtohandleWAND.However, notallquerieshavemandatoryterms. Inthiscaseweapplyasort-basedmethodoraLoserTreebasedone according to the WAND threshold. The experimental results demonstrate that ourframework is much more e?cient than the previous method, and also very robust.