Several years ago I was able to use IBM’s Watson Personality Insights to analyze my personality based on how I use Twitter (see IBM’s Watson Artificial Intelligence Analysis of My Personality). It contained both positives (You are philosophical: you are open to and intrigued by new ideas and love to explore them) and negatives (You are shrewd and somewhat critical), but overall I would say was reasonably accurate.
Well, IBM Watson Personality Insights is no longer publicly available ( I suspect it was a little too good?) but I did find another tool! Fresh out of the University of Cambridge Psychometrics Centre, we have the Apply Magic Sauce social media analysis program.
Some of my results are posted above, and,to my pleasant surprise, in the digital space I am only 34 years old—roughly a quarter of a century younger than YIRL (years in real life). (My wife is pleased about this finding; she says it explains a lot!)
And my Big 5 Personality? Well, that looks quite accurate. I do tend mostly to hover around the average, except I am quite contemplative (not particularly interested in engaging with the world), and I do indeed drift towards the liberal and artistic end.
So, is this all meaningless, just some Natural Language Processing fun and games? Perhaps, but consider: this was a digital phenotypic profile, based on only 777 tweets and 20,499 words. Imagine what Amazon, Facebook, and Google can and are doing, with not just this same set of tweets, but with thousands of other things I have done online? Imagine the value of that digital phenotype to health insurance companies, potential future employers, and advertisers?
Anyway, this tool from University of Cambridge just shows the tip of the iceberg of what AI and Big Data can do in terms of creating a digital phenotype of a person, so for this reason alone, it’s worthwhile playing around with it.