VisualTextRank: Unsupervised Graph-based Content Extraction for Automating Ad Text to Image Search

TL;DR

VisualTextRank is an unsupervised graph-based method that improves ad image retrieval by extracting relevant keywords from ad text, leveraging similar ads and semantic embeddings, leading to better search and onboarding outcomes.

Contribution

The paper introduces VisualTextRank, a novel unsupervised approach that enhances keyword extraction for ad image search by integrating semantic similarity and advertiser-specific biases.

Findings

Achieved 11% accuracy improvement over biased TextRank.

Increased stock image search usage by 28.7%.

Boosted advertiser onboarding rate by 41.6%.

Abstract

Numerous online stock image libraries offer high quality yet copyright free images for use in marketing campaigns. To assist advertisers in navigating such third party libraries, we study the problem of automatically fetching relevant ad images given the ad text (via a short textual query for images). Motivated by our observations in logged data on ad image search queries (given ad text), we formulate a keyword extraction problem, where a keyword extracted from the ad text (or its augmented version) serves as the ad image query. In this context, we propose VisualTextRank: an unsupervised method to (i) augment input ad text using semantically similar ads, and (ii) extract the image query from the augmented ad text. VisualTextRank builds on prior work on graph based context extraction (biased TextRank in particular) by leveraging both the text and image of similar ads for better keyword extraction, and using advertiser category specific biasing with sentence-BERT embeddings. Using data collected from the Verizon Media Native (Yahoo Gemini) ad platform's stock image search feature for onboarding advertisers, we demonstrate the superiority of VisualTextRank compared to competitive keyword extraction baselines (including an $11\%$ accuracy lift over biased TextRank). For the case when the stock image library is restricted to English queries, we show the effectiveness of VisualTextRank on multilingual ads (translated to English) while leveraging semantically similar English ads. Online tests with a simplified version of VisualTextRank led to a 28.7% increase in the usage of stock image search, and a 41.6% increase in the advertiser onboarding rate in the Verizon Media Native ad platform.