Any technology based upon semantics, should be capable to identify the meaning and context of the words on the page, in order to determine the appropriate advertising. This is an advancement to the current contextual model where the simple identification of keywords is seen as sufficient to represent the context of the entire page.
The semantic solution allows for a broad ranging, often human developed, taxonomy of advertising categories. The taxonomy categories will contain all the related terms and phrases articulating a specific subject, enabling the categorization of a page as being about a specific subject.
The machine learning technologies are often based upon concepts such as the naive Bayes classifier or support vector machine. These technologies require 'training' to assimilate data to lead to effective categorization but are scaleable and have multilingual capabilities.However, in many cases, the use of the phrase semantic in describing a product is in fact a misnomer as the products are simply extensions to the current contextual model and provide an ability to "identify" the high value keywords on a webpage.
The advantages of using semantic advertising are the capabilities to identify the entire content of a webpage and extract all of the core themes from within the content, enabling the advertiser to choose the most commercially relevant theme. A further advantage is the ability to perform word sense disambiguation, i.e the ability to understand the correct sense of the words used in order to avoid misplacements of ads. To use the previous example, semantic advertising would be able to differentiate between pages discussing zoos versus cars.
Deploying semantic technologies for advertisement can serve a whole set of other benefits as well. Semantic advertising can also recognize sentiment, for example, if a web page is discussing a topic in a negative or positive manner. Therefore, an ad for the next US Democratic presidential candidate would be posted on a web site with a positive opinion of the person.In addition, semantic advertising products such as ad pepper's iSense products, provide a capability to filter potentially objectionable content such as adult, gambling, drugs and oter such controversial themes.
Semantic advertising is a valuable technology because it increases an advertiser’s return on investment and allows an Internet user to more easily locate an appropriate product or service. Semantic advertising also allows analytical functions that provide data to track the effectiveness of an advertising campaign.
WIPO ASSIGNS PATENT TO MICROSOFT FOR "SEMANTIC ADVERTISING SELECTION FROM LATERAL CONCEPTS AND TOPICS" (AMERICAN INVENTORS)
Aug 12, 2011; GENEVA, Aug. 12 -- Publication No. WO/2011/097067 was published on Aug. 11. Title of the invention: "SEMANTIC ADVERTISING...