Impedance coupling in content-targeted advertising

Abstract

The current boom of the Web is associated with the revenues originated from on-line advertising. While search-based advertising is dominant, the association of ads with a Web page (during user navigation) is becoming increasingly important. In this work, we study the problem of associating ads with a Web page, referred to as content-targeted advertising, from a computer science perspective. We assume that we have access to the text of the Web page, the keywords declared by an advertiser, and a text associated with the advertiser's business. Using no other information and operating in fully automatic fashion, we propose ten strategies for solving the problem and evaluate their effectiveness. Our methods indicate that a matching strategy that takes into account the semantics of the problem (referred to as AAK for "ads and keywords") can yield gains in average precision figures of 60% compared to a trivial vector-based strategy. Further, a more sophisticated impedance coupling strategy, which expands the text of the Web page to reduce vocabulary impedance with regard to an advertisement, can yield extra gains in average precision of 50%. These are first results. They suggest that great accuracy in content-targeted advertising can be attained with appropriate algorithms.

Using landing pages for sponsored search ad selection

“…For instance, Ribeiro-Neto et al [25] examine the use of vector space model and cosine similarity as a ranking function for content match ad retrieval. To resolve the vocabulary mismatch, the triggering page (used as a query) is expanded by features from related pages.…”

“…To resolve the vocabulary mismatch, the triggering page (used as a query) is expanded by features from related pages. The proposed method in [25] is particularly suitable for content match, but the application to sponsored search is not straightforward due to lack of the triggering pages in sponsored search.…”

“…Although most of previous research for ad retrieval did not utilize landing pages for ad-side expansion, some (e.g., [25,24]) experimented with augmenting the ad with the entire landing page to improve the content match. In contrast, we perform a more focused study of landing pages, contrasting a number of in-context and out-of-context extractive summarization techniques.…”

“…In contrast, we perform a more focused study of landing pages, contrasting a number of in-context and out-of-context extractive summarization techniques. In addition, our study is first to explore the use of landing pages in the context of sponsored search, where the query is much shorter than that of content match studied in [25,24].…”

“…The advanced match problem corresponds closer to the web search problem. Recent advanced match approaches use search techniques for ad selection by evaluating the query over a corpus of documents that are created from the ads [7,25]. One of the key difficulties in this ad retrieval approach is that the ads are much shorter than documents in most other search applications [25].…”