| NetSeer

stochastic modeling

More than you’d think.

NetSeer’s own co-founder and chief scientist, Dr. Vwani Roychowdhury has the inside track on the data – and yes, it’s far less morbid than it sounds.

Vwani’s research was recently published in a co-authored study, Stochastic Modeling of a Serial Killer, that analyzes the time-pattern activity of a serial killer using predictive algorithms to anticipate a next move. But what does this have to do with ad-targeting strategies?

Vwani chimed in: “In this research, we first determined the patterns in the time interval between consecutive homicides of serial killers and then proposed a brain activity model that might lead to such behavior patterns. What’s at the heart of this is the ability to capture intent and underlying dynamics by observing human actions. It takes advanced pattern-recognition algorithms to uncover these triggers and develop a predictive engine.”

“Think about it this way: the same kinds of patterns appear in other far more innocuous human activities, such as time intervals between answering emails. Or, visiting a favorite website such as a fashion e-tailer, and potentially purchasing comparable items. One of the findings from this study revealed that if an action is based on an ‘urge’ that needs to be satisfied, then people who have committed an ‘act’ – such as buying a pair of shoes – are more likely to repeat the act again (i.e., buy another pair of shoes) within a short time interval.”

“This shines an interesting light on why retargeting can be incredibly effective, if executed properly. Lookalike audiences and retargeting, even for items a user might already have purchased, provide a mechanism to tie the ‘urge’ with the ‘act’. It’s the power of a really relevant suggestion.”

Vwani will take a deeper dive into the ‘math behind the motives’ at next week’s Predictive Analytics Conference #PAWcon. On Wednesday, June 18th, he’ll tackle consumer cognitive modeling during his session Media and Murder: By the Numbers.

We’ll see you in Chicago!