OracleLabs researchers have developed over the last 15 yrs an adv ML pattern recognition innovation called the Multivariate State Estimation Technique (MSET2), for prognostic surveillance of time series signals from dense-sensor IoT industries for predictive maintenance applications, with high sensitivity for proactive warnings of incipient anomalies, but with ultra-low false- and missed-alarm probabilities (FAPs and MAPs). This TechCast will demonstrate that MSET2 possesses significant advantages over conventional ML algorithmic approaches, including NNs, AAKR, and SVMs. MSET2 advantages include: higher prognostic accuracy, lower FAPs and MAPs, and much lower overhead compute cost, a crucial differentiator for real-time dense-sensor streaming prognostics. MSET2 brings significant value add for Oracle streaming analytics as well as "in-DB anomaly discovery" in the fields of Utilities, Smart Manufacturing, Oil&Gas, Commercial Aviation, and of course for IT assets in datacenters.
By registering for this TechCast you give permission for your name and email address to be shared with the presenter and for BIWA User Community so we can inform you of future TechCasts and conferences of interest.