Web Page Summarization for Just-in-Time Contextual Advertising

Abstract

Contextual advertising is a type of Web advertising, which, given the URL of a Web page, aims to embed into the page the most relevant textual ads available. For static pages that are displayed repeatedly, the matching of ads can be based on prior analysis of their entire content; however, often ads need to be matched to new or dynamically created pages that cannot be processed ahead of time. Analyzing the entire content of such pages on-the-fly entails prohibitive communication and latency costs. To solve the three-horned dilemma of either low relevance or high latency or high load, we propose to use text summarization techniques paired with external knowledge (exogenous to the page) to craft short page summaries in real time. Empirical evaluation proves that matching ads on the basis of such summaries does not sacrifice relevance, and is competitive with matching based on the entire page content. Specifically, we found that analyzing a carefully selected 6% fraction of the page text can sacrifice only 1%-3% in ad relevance. Furthermore, our summaries are fully compatible with the standard JavaScript mechanisms used for ad placement: they can be produced at ad-display time by simple additions to the usual script, and they only add 500-600 bytes to the usual request. We also compared our summarization approach, which is based on structural properties of the HTML content of the page, with a more principled one based on one of the standard text summarization tools (MEAD), and found their performance to be comparable

Towards context-aware search with right click

“…To minimize latency and communication costs, summarizing the content of the Web page is often done prior to performing a match search on the advertisements. In [1], the role of page components in crafting short but informative page fragment that serve as an alternative description of the entire page is studied. Their experiments show that the combination consisting of page URL, referrer URL, title, meta-data, and headings serves as a good page summary.…”

SmartAds

“…Note that our goal is related to recently proposed techniques for "just-in-time" contextual advertising in dynamic web pages [5,20]. These techniques aim to deliver relevant ads to dynamic web pages based on useful signals extracted from them during runtime.…”

“…In general, these techniques rely on various properties of web pages and are not directly applicable for mobile apps. For example, [5] uses page URL as well as referrer URL, which do not exist for mobile apps. It also sends short excerpts of the page to backend server, and hence violates privacy.…”

“…Therefore, if the set of bidding keywords that we use for local pruning 4 We discuss in Section 6.2 how the remaining 10% ads actually get served to clients. 5 Deletion is supported in counting Bloom filter that has more memory footprint and therefore we do not use this. changes dramatically, the client needs to download the entire Bloom filter from the server.…”

“…Inspired by them, SmartAds further employs features that are specific to mobile application UI in order to improve the quality of extracted keywords. "Just-in-time" contextual advertising dynamically extracts keywords and other signals from dynamic web pages for selecting relevant ads [5,20]; however, as mentioned in Section 2, they use web page-related features and are not directly applicable to mobile apps. Beyond keywords, several 'deeper' features have been employed for delivering relevant ads, including semantics mined from web pages (e.g., topics) [7] and user interactions in browsing [9].…”