AI expands digital marketing

Working via a knowledge transfer partnership (KTP) with Crimtan, the project - which has a value of £125,372 - will be led by Professor Miltos Petridis and supervised by Dr Roger Evans, both from the university's School of Computing, Engineering and Mathematics.

It will use computational engineering and artificial intelligence to develop the new technologies, fostering the company's commitment to responsible, highly effective advertising. The primary focus of the academics' work will be enabling brands to benefit from AI and advanced mathematics research via a new portfolio of real-time, cluster-targeting advertising systems that protect user privacy. Derived from the agency's statistical data and human capital, the emergent advertising technologies will automate online advertising processes and decision-making by mirroring human expertise and experience.

At a basic level, the technologies will make radically more intelligent and vastly accelerated decisions, such as which specific creative from a brand's campaign should be displayed to a particular user, in which format and when.

Evaluating negative responses

Of particular note is the use of negative modelling techniques. At present online ad campaign decision-making is primarily governed by user actions, such as clicks, online purchases, hover-overs and the like. To date, the value of inactions has not been effectively incorporated into online ad systems. Negative modelling techniques will help to define decision-making rules based on inactions, non-occurrences and non-events. At a basic level, this relates to data regarding users not clicking on an ad, not purchasing after entering an online store, or not hovering over an expandable rich media advert. The resultant new systems will limit wasted ad displays, potentially giving new relevance to impression-based online advertising metrics, such as CPM - the cost, per 1000 people reached, of buying online advertising space.

Dr Evans said: "Forward thinking companies like this are looking to technological innovation to improve responsible online advertising. In particular this partnership will be investigating how AI can be used in online marketing targeting systems using negative and lookalike profiling as well as other innovative techniques.

"Advanced mathematics has helped the Internet to evolve into what it is today but there's room for improvement. We will be using advanced mathematics to define logical rules that underlie both predictable and surprising behaviour of Internet users, brands and online advertising experts. The partnership will help to break new technological boundaries for the digital marketing industry. These technologies will offer vastly improved services for innovative advertisers, while enriching the online experience for Internet users."

Evolving digital marketing

Paul Goad, CEO of Crimtan added, "The digital advertising industry is increasingly driven by data and regulation is growing increasingly severe. Digital marketing needs to evolve and use new techniques and technologies to adapt to customer concerns. These new technologies are inspired by our commitment to the most effective and responsible online marketing.

"The company sees this KTP as a further step that will enhance the ongoing efforts of its large development team to keep the company at the cutting edge of this extremely fast paced industry. The university's world class expertise will help the company to develop the kind of technology that will enable the digital marketing industry to innovate, evolve and improve."

Over the course of an intensive two year programme the KTP Associate will work to develop, test and implement the range of new technologies.

Professor Petridis said, "Having been involved with KTPs before joining the university, I know that they are a great vehicle for working with companies and feeding that experience back into teaching and research.

"I have also witnessed the catalytic-transformational effect a KTP project can have on companies such as this, their growth, business portfolio and confidence. We are looking forward to the opportunity to apply our expertise in data mining and modelling in the company's highly sophisticated, large-scale, real-time web delivery environment."