Introduction

1.1  People behave in predictable ways and what they are likely to do or how they might turn out in the future can with some degree of accuracy be predicted by analysing what they have done or experienced in the past.

1.2  Big data analytics combined with the processing power of HPC (High Performance Compute ) enables the accuracy of prediction to be enhanced. The assessment, triangulation and comparison of multiple streams of data are now possible in a way that until recently was not practical. By using machine learning and big data processing, we can better predict future need and demand for public sector services.

2.   Relatedness Technology

2.1  Relatedness technology creates a database of connectedness between people, places and events. This database of relatedness enables the prediction of how people may behave or react in certain situations.

2.2  It not only analyses past events but also assesses the strength of the links between correlated events. This will enable better prediction of demand for services.

2.3  Relatedness Technology takes the social interactions and historical events concerning a person to enable the prediction of future events and it has four distinct features

·       It creates a database of the connections

·       It can assess which links are the strongest

·       It can find future connections

·       It can offer some predictions about the significance of future connections.

3.   Deployment in a Schools and Child Protection Setting

3.1  Early intervention in the lives of children, who might otherwise fail to thrive, could be a cost-effective way of preventing higher future costs. Identifying those children can be done by using an algorithm which is applied to a range of data sources

3.2  Children who fail to thrive and who go on to lead lives which require significant state intervention are likely to have experienced a number of adverse childhood experiences (ACE)  or in other ways have come to notice for health, behavioural or other reasons.

 3.3  By identifying the nature, scope and timing of those ACEs and other negative factors in those children who have failed to thrive and building those into the algorithm it will be possible to identify children who will fail to thrive in the future by algorithmically analysing their records.