Last week we previewed this in “How Machine Learning APIs are Being Used to Predict Startup Success.” Can there be a quantifiable way to hedge investors’ risk and ensure they are betting on the right horse? According to the startup “jury” algorithm PreSeries, it’s mathematically probable to predict which startup is most likely to succeed and that startup is Novelti. This startup which uses online machine-learning algorithms to convert Internet of Things sensor data into real-time intelligence, machine learning and pattern recognition was predicted to be successful with an 87 percent likelihood.
Novelti beat out four other predictive analytics and artificial intelligence competitors—Intranetum, Emotion Research Lab, Datatrics and restb—at the PAPIs Connect conference for machine learning and predictive APIs. Using a voice-command device, PreSeries listened to each startup’s four-minute pitch and then asked each a series of questions. This was combined with data from 27,000 past venture funding rounds in order to determine the winner of startup with the most promising future.
“As a serial entrepreneur myself, I have always felt that traditional early-stage venture funding has been very slow to incorporate data-driven decision making,” said Francisco J. Martin, president of PreSeries, and co-founder and CEO of BigML REST API for programmatic machine learning, which partnered with Telefonica’s Open Future_ to develop PreSeries. “The results speak for themselves as founders keep suffering from the lack of objective and constructive feedback on where their companies stand, and the investors have not been able to improve on their hit rates in decades.” He went on to say how he felt this was an example of how machine learning can be used to “disrupt this old-fashioned approach.”
Novelti will join the next Wayra class at Telefonica’s startup accelerator and get to directly compete in the AI Startup Battle in Boston at year’s end.
What are quantifiable ways to determine success? Is an algorithm a better judge of startups? Tell us what you think below or tweet to @ProgrammableWeb and #PAPISConnect!