Full Path URL content classification is superior to base domain classification.
Here are some examples:
Yahoo.com – This is considered a base domain, which is a URL that cannot be shortened any further.
Health.yahoo.com – This is considered a sub-domain. It is a URL which consists of the base domain along with a prefix. In this case it’s “Health.”
Health.yahoo.com/diet-fitness – This is considered a sub-path or sub-page. It is a URL which consists of the base domain or a sub-domain along with a suffix. In this case it’s “diet-fitness.”
The content on above-mentioned URLs is different and when queried to zvelo’s categorization engine, they yield varying category values. Yahoo.com should be categorized as a “portal site,” health.yahoo.com as “health-other” and health.yahoo.com/diet-fitness as “diet” and “exercise.”
Other great examples are Blogspot.com URLs, which host a wide variety of professional or personal blogs.
Blogspot.com – A base domain and should be classified as “Personal Pages & Blogs” and “Web Hosting, ISP & Telco.”
Nookandpantry.blogspot.com – A sub-domain URL that should be classified both as “Personal Pages & Blogs” and “Food – Other.”
Imagine an online advertising platform provider needing to place highly targeted ads for a health and fitness company. An ad placed at the base domain Yahoo.com may not be the most ideal or relevant to the site’s readers, whereas an ad placed on the sub-domain health.yahoo.com may produce a more attractive return on investment for the said advertiser.
Other real-world scenarios may include a network security company or parental controls software maker needing to detect and block end-user access to objectionable web content. It has been observed that Blogspot.com blogs are riddled with pornography, hate, violence and other unsuitable content. For this use case, content classification at the sub-domain (adultsite.blogspot.com) instead of the base domain alone (Blogspot.com) is a strict requirement.
zvelo artificial intelligence-based contextual categorization engine, backed by stringent human quality assurance review, is capable of differentiating millions of full path or base domain URLs so that the most ideal category value is attained. This AI-human merged model makes for the most comprehensive categorization services for the ad tech, big data analytics, network security and other markets.