US Patent 9058623 Semantic tagged ads

A scraping mechanism may download a web page from a retailer and extract information for each item being offered for sale. The scraping mechanism may parse the published HTML code to identify items for sale, along with other parameters that may be presented in the web page. From the web page, a data structure may be created that represents the items for sale, and the data structure may be used to generate advertisements for the retailer, add the items to a search engine, or for other uses. In some embodiments, the retailer may include hints, tags, or other annotations in the HTML code to facilitate the data capture. The web page may be a standard public page or may be a private page accessible to the scraping mechanism only.