Findthem.

Everyday people go missing across the world. We process data from a huge number of sources to provide tools to help find them quickly.

Everyday hundreds of people go missing across the world. 

The  Missing Persons Data report ('http://missingpersons.police.uk/en/resources/missing-persons-data-report-2014-15) gives the following information:

"In the 2014/15 year, across the English and Welsh police forces there were 282,066 missing person-related calls, of which 210,632 were created as incidents, relating to 129,046 individuals. Police Scotland received 39,926 missing person-related calls, an increase of 9% compared with the 2013/14 year.

In 2014/15 just over one third of the incidents were attributable to repeat missing. This demonstrates that in line with the previous year, there is a slight increase in repeat incidents"

Ascertaining the locations in which to search for a missing person can often be difficult and often relies on data from a variety of different sources but at present there isn't a tool in order to do so. 

What does findthem do?

In order to find someone quickly data from a variety of different sources must be gathered and processed in order to ascertain the missing persons possible locations. Manually sifting through this information is time consuming. It can be difficult to quickly ascertain just how far away the person could have gone during the time since they went missing. CCTV footage can be slow to get since it often relies on police forces asking shops rather than them being alerted by an automated service. 

We need funds in order to undertake further research, hire a developer to start to put together the service, put together promotional material and start a pilot with a number of different police forces across the UK. 

How does it work?

Findthem takes data from a large variety of different sources such as transport information (train departures, current traffice reports etc), banking, CCTV, hospital data, accident reports, sightings, manually added data (age, sex, time of going missing, social media links, car access, financial situation) to process and then display on a map a highly detailed view of the possible search locations with the relevant information. 

This view takes into account housing boundaries, road access, traffic, transport departures and other such information to ensure that the information shown is accurate. With the passing of time as the search continues, the map is updated live as the circumference that needs to be searched increases or the possible transport links change.  

The Service automatically checks given Social Media links for any new posts, tweets or messages and provides alerts if it notices anything on these accounts. Automating processes which would have traditionally been undertaken manually both frees up police resource and allows for immediate action to be taken should a report be received. 

In addition to the map, alerts are sent to relevant groups (who are signed up) of people (bars, schools, hotels, hospitals, shops) of the missing person, along with a photo and police log so any sightings can be reported. If required social media is also used to alert other users of the missing person. 

When someone is located successfully the groups originally notified as again sent a message to let them know the person has now been found. 

Example

The example below shows a very quick mock up of the type of interface that could be displayed.  Here a child of 7 years old was shopping with his family when he went missing. The mother alerts the police, who immediately send out an auto alert to all those signed up (shops, council workers, traffic wardens etc.) to the scheme. Details taken from the mother are fed into the Service and it immediately starts to track the possible locations. 

In the example above the boy has been missing for sometime, yet there have been a number of possible sightings both reported by the users who were alerted. A simple alerts gives the option to reply with no sighting, possible sighting or confirmed sighting. The boy often gets the train with his family and so train information of recent and upcoming depatures is also displayed as well as their next stopping locaton. Further information can be found out about these trains if needed. The system will keep a live track of possible search areas in neighbouring towns in a similar way as shown on the map above.  

As well as trains the service could keep track of buses, trams, ferrys and other modes of transport. The service would use a weighting to decide how likely it was the person was able to catch each mode of transport. Colours indicate which modes of transport are more likely. 

Given the recent sightings from a boy to the South West of the point that he went missing the system starts to suggest the most obvious locations to search. As the search goes on, the search area (marked in green) changes and responds to this new information as time goes by. In the example above it's likely the search area will expand to the South West. 

In order to provide a more accurate picture of the search area dead ends, boundaries, fences, walls, traffic etc are all mapped. This gives a much greater level of detail about how far a user may have travelled on foot. 

Discounting information

If information is known about the missing person that makes any tracking data irrelevent than that information can be discounted. For example, if it is known that the missing person does not drive then potentially we can ignore traffic and driving distance layers to focus the search more closely. 

Repeat Incidents

In 2014/15 just over one third of the incidents were attributable to repeat missing. If someone regularly goes missing the data can be used overtime to build up patterns about when and how they go missing. Do they always go missing from a certain location? When they go missing do they always go to the same location? How long do they go missing? Do they return home after a certain amount of time on their own accord?

Thank you for your time.