There are three kinds of computer-assisted intelligence: augmented intelligence from individuals, collective intelligence from networking, and artificial intelligence from computers that mimic human intelligence. These three are merging, as well as the Internet and our minds, which also means that the kind of Journalism that we have now is influenced by these three intelligences. We see that evident in crowdsourcing, reader comments, etc.
In this article, we will focus on the section about Behavioral Targeting and Journalistic Content, of the article written by Noam Lemelshtrich Latar and David Nordfors. Get the pdf of the article here: The Future of Journalism: Artifical Intelligence and Digital Identities.
According to Aho Williamson, behavioral targeting is the ability to deliver ads to consumers based on their behavior while viewing web pages, shopping on-line for products and services, typing keywords into search engines or combinations of all three. This new form of marketing first gained attention during the late nineties, and three major online companies, Microsoft, Yahoo and Google use it. Google aims for targeted, and therefore more relevant ads, increasing the value of these ads. The company uses AdSense, which targets ads based on the content of the website from which these ads appear, and AdWords, which targets ads based on the inputted keywords. Furthermore, Google also targets ads in Gmail accounts by examining the spam messages and basing the targeted ads from the key words. Annotation of video and image content is not yet available, and behavioral targeting is for text as of now.
Behavioral Targeting in Social Networks
With social networks, users voluntary share personal information about themselves, including interests, demographics, friends, personal videos, pictures and texts, among others. It’s no surprise why behavioral targeting has now grown increasingly in these networks.
An example company which uses behavioral targeting in social networks is 33Across.com. This company studies behavior patterns among consumers in the social network setting. They look for members of that network which have the most influence among his or her friends, identified through message dynamics. Relevant marketing ads are then created. Tacoda Systems and Reverence Science also use targeting on social networks.
Consumer profiling through behavioral targeting is the subject of continuous discussions because of the serious privacy concerns it entails. Among the advocates of raising awareness to several implications, including ethical, of consumer profiling is Tim Berners -Lee himself, inventor of the world wide web. But there are also privacy studies showing consumers are willing to be given enhanced content delivery.
Managing Digital Identities
A lot of industries are investing to improve finding techniques to obtain digital identities. User profiles are gathered and organized to follow a world wide standard in what is known as federated identity management. An example of this standard is SAML2, or Security Assertions Markup Language 2.0. This standard is used in several academic institutions, financial organizations and the electronic government of America. In the near future, everyone who surfts the web will have a personal digital identity, describing one’s personality through investigation of data obtained by “robots”. Furthermore, digital identity and social networking is a powerful and dangerous combination, as they would then allow profiling which includes the professional and social connections of an individual, without their awareness.
Behavioral Targeting AI Engine
Using Digital Identities and Journalistic Content, a behavioral targeting AI engine is created. First, it will analyze all journalistic content by a user, and each content will be tagged or annotated automatically. Second, these tagged content and the digital identity of the consumer will be sent to the Assessment Rule Engine, which will then send relevant ads to that consumer.
Third, as the consumer interacts with these ads, his or her behavior and attention towards those ads is constantly monitored. Fourth, the Learning Engine will analyze the feedback of the consumer to adjust for a better description of the behavior of the consumer. Thus, new content will be sent. Fifth step involves updating this progress to a Personal Memory database, and the fourth and fifth steps will keep going indefinitely for accurate consumer interests and choices.