An Empirical Study of Generating Texts for Search Engine Advertising

Abstract

Although there are many studies on neural language generation (NLG), few trials are put into practice, particularly in the advertising domain. Generating ads with NLG models can help copywriters in their creation. However, few studies have adequately evaluated the effect of generated ads with actual serving because this requires a large amount of training data and a particular environment. In this study, we demonstrate a practical use case of generating ad-text with an NLG model. Particularly, we show how to improve the ads’ impact, deploy models to a product, and evaluate the generated ads.