We are inundated by data, in our professional and personal lives. How do we make sense of it, extract the insights we need? An image is worth a thousand words, but can you visualize a complex dataset with multiple variables and scores of datapoints? This is what Leo Meyerovich of Graphistry set out to do, building on his PhD research at Berkeley. Graphistry uses Artificial Intelligence to trace patterns and correlations, helping the analyst in her investigation of the data.
In some cases, we are in fact talking of true investigations, the stuff of cloak and dagger movies. In this episode you will hear about financial fraud, election influencing, drug and human trafficking. Leo describes how AI-driven data visualization helps uncover shady webs of unsavory communications and illegal money flows, even as the criminals carefully try to cover their tracks. The same techniques can help a company better understand the behavior of its customers and how to reach them and serve them more effectively. You will also hear how Graphistry can help capture the experience and tricks of senior analysts and make them available to younger recruits—an important feature as rapid staff turnover becomes a more pressing challenge for many organizations.
At the end of the episode, Michael and Marco discuss future potential applications and debate whether future data analysis will still have a ‘human in the loop’ or evolve towards a pure black box approach.