ende
  • +44 (0)20 7060 6990

Our technology

Developed over 10 years of academic and commercial research, our software, algorithms, platforms and technology are the most advanced in automated predictive analytics globally.

Automating the predictive analytics process

Predictive Analytics used to be a project, now it is a product. The automation and configuration of steps makes our software platform truly able to be used by a business user with minimal necessity of involvement of IT or data science in the process. Crucially note the simplifications for example data are not cleansed, and the required algorithm and any other processing are determined by profiling the data available and the end-user defining the problem.

Our predictive analytics software is able to address these inhibiting challenges so that they can be used reliably by both data scientists and non-data scientists alike. Adoption and deployment of these tools and applications will greatly enhance the ability of businesses to move quicker and more cost-effectively by dramatically lowering the cost barriers to predictive analytics, and turning it into a ‘product’ rather than a ‘project. It is truly democratizing predictive analytics throughout organizations.

WA_DIAGRAMS_V2-02
How it works

How it works

Step 1: Data are ingested in their raw form into the proprietary Big Data Platform (known as “A3”). Connectors built for most standard databases and storage infrastructure.

Step 2: AIR (Automated Information Retrieval) is one of the proprietary suites of algorithms which rapidly and automatically transforms the data based on the workflow only. Data don’t require cleansing. AIR removes the intensive processes typically 50%-80% of a data scientists time.

Step 3: The data science workflows are pre-built in the core A3 stack. Some of these contain themselves further proprietary algorithms and are themselves unique. They can be easily combined and configured to address business use cases.

Step 4: The Early Warning and Prediction/Prevention enables continuous real-time analysis. The data science workflows can be easily configured and combined to address business use cases e.g. Customer dynamic segmentation
(Explanatory Analysis + Classification + Association Rules), Fault prediction/root cause analysis (Classification + Clustering + Anomaly Detection), Fraud prediction, Process optimization (Explantory Analysis + Anomaly Detection).

AIR (Automated Information Retrieval)

  • AIR automatically takes heterogeneous data feeds and naively exhaustively transforms and prunes to provide structured data for pattern mining
  • No prior modelling required
  • Dictionary creation is automated output – not input. Benefits are speeding up process and learning syntax used, not guessing
Web

© 2017 Warwick Analytics. All rights reserved. Registered in England & Wales. Number 07724630. Registered address 35 Kingsland Road, London, E2 8AA. VAT 120435168.

Warwick Analytics