710655Automated Information Retrieval (AIR)ClosedGRD Proof of ConceptInnovate UKThe boom in big data and predictive analytics is well documented. Yet despite all the hype about the potential insight and unlocked value, the vast majority of data are unstructured (at least 80%) including ‘Internet of Things’ (IoT) type data, web, text, audio and video. This is only set to grow both in volume and proportion. The state-of-the-art predictive analytics techniques are only able to function within this data by using bespoke and specific pattern recognising tasks (called “Information Retrieval” or “IR”). There is currently no holistic IR technology which can analyse any common unstructured dataset. If there were, then e.g. people’s purchasing behaviour could be predicted from their mobile phone signal; a salesman’s performance could be predicted from their emails; a patient’s health could be predicted from their doctor’s notes combined with their home utility usage. And all without elaborate bespoke models. However it is, as it sounds, a hard problem at the cutting edge of both data science and computer science. Yet Warwick Analytics has successfully created a PoC for automated IR within a specific domain (aerospace maintenance) and a Smart PoM for market demand, and is looking to extend that work to build a PoC of the world’s first holistic, Automated IR or “AIR”.