Machine Learning Platform with the capability to predict crimes

Currently available in a closed alpha Yellow Submarine today announced that it's machine learning platform now has the capability to predict crimes.
BEDFORD, U.K. - June 9, 2019 - PRLog -- To coincide with the launch of the software, Yellow Submarine has also signed a partner agreement with IBM and the software is available through them

Yellow Submarines CEO Nash Ramadhen said, " Our Machine Learning Platform is capable of predicting many different outcomes for the emergency services. We are currently trialling the software in open alpha and invite anyone interested in working with us to contact us. We are also working on developing models to predict fires and also specific crimes that can be attended by ambulances."

Yellow Submarines machine learning platform isn't just for emergency services though. It can be used for many industries, such as finance and retail. The idea for Yellow Submarine came from both Nash the companies CEO and Justin the companies CTO after successfully being one of the winners in the JLAB (John Lewis Lab) with footfall prediction software for bricks and mortar stores.

"I wanted to build a machine learning platform that doesn't use loads of historic data. Looking too far into the past is a bad idea. This is why models built using neural networks trained on such data can become overfitted", said CTO Justin Staines, "our software looks at small and smart data - from the big data. We look for anomalies and using a combination of GLM and Clustering can predict hours, days or even months in advance with greater accuracy than other types of modelling."

About Yellow Submarine

Yellow Submarine is a privately owned company based in Bedfordshire. Its intellectual property is protected by patents GB1819646.9, GB1819645.1, GB1819644.4, GB1819643.6, GB1819642.8, GB1819641.0
Source: » Follow
Email:*** Email Verified
Location:Bedford - Bedfordshire - England
Account Email Address Verified     Account Phone Number Verified     Disclaimer     Report Abuse

Like PRLog?
Click to Share