Monthly Archives

April 2015

Track Group Location Analytics Offers Deep, Broad, Predictive and Layered Perspectives

By | News

By: John Kirtland, Track Group U.K.

We have all seen the rapid increase in data creation – much of which has location information as an integral part. Developments in toolsets and algorithms for the analysis of large volumes of location based data now enables greater context to be derived in Pattern of Life (PoL) analysis… answering more questions in less time and with the benefits of collaborative work processing.

The tools to view the data have typically been delivered through map-oriented or data-oriented visualisation efforts. The reality of these tools is that they represent a visualisation into the raw data – leaving all of the interpretative analysis of what the data actually represents up to the user. Users are now looking for automated answers that reveal key patterns that are represented in the data but that are not evident through pure visualisation.

The development of advanced generation toolkits in response to increased data provision and in-depth questioning are now available and producing good results. Agencies are reporting improved intelligence results with reduced Analyst processing time.

Location Analytics

To obtain intelligence from all of these analytical stages, one must be able to effectively apply four treatments to location-based surveillance data:

  1. Deep Analyzing several months of data, from a number of sources, for a suspect revealed his location to match a number of other crimes that he had not been seen as connected with.
  2. Broad The analysis of data relating to 8 targets gathered over 3 months was completed in 11 minutes. Comparing each target against all the others identified previously unrecognized common meeting and visit locations connecting two separate groups and resulted in a modified approach to their surveillance.
  3. Predictive Location data for a target with complex life patterns identified their home, work and other regular locations in 90 seconds, an activity that had previously taken weeks of physical surveillance of this unpredictable pattern.
  4. Layered Following apprehension of a gang leader other suspect tracks were analyzed with influential gang members being identified by their response behaviors captured through phone records and social media activity. The same analysis also eliminated other traces from the subsequent round of pick up activity.

Deep – rapidly analyse 100,000’s data points

The analysis of large data sets has been made easier and can be processed faster to automatically provide a Deep Life pattern.

There has been a large increase in the amount and variety of data presented to intelligence and surveillance teams. First and second generation toolsets were unable to make best use of this data quickly resulting in analysis commonly being applied to recent activity only.

Toolsets provide answers to commonly asked questions:

  • Expand awareness of the location of interest with certainty of the previous and subsequent locations, along with the routes used. This allows plans to be developed that can reduce the cost of surveillance activities and enhance the deployment of personnel and technology.
  • Bring in a greater volume of data to identify other locations of interest previously not identified through regular use or abnormal use. Such locations do not normally reveal themselves with superficial analysis.
  • Identify cycles of behavior previously not identified nor planned around. Longer term cycles of activity can only be identified when analyzing greater periods of collected data. Third generation tools can analyze this volume and also conduct greater contextual analysis in doing so. 


Broad – compare across multiple targets

Processing multiple parallel sets of data for multiple targets is providing a Broad Life pattern within a matter of seconds.

The comparison of a single target against a small number of others (1:N) was a useful feature of the better second generation toolsets. Third generation tools have advanced to enable N:N analysis at the push of a button and without the need for a supercomputer.

With the added advantage that the analysis can be performed across large data sets a Broad Life pattern covering these three key features can be easily delivered in a single review of the data.

  • Review common locations across many targets, and whether their attendance at these locations is also common in time, to determine meeting places.
  • Perform a similar review where the location attendance is not common in time to identify drop locations or safe houses.
  • Review tracks for multiple targets to identify previously unknown connections between individuals or groups.

Predictive – where next? and when?

With an enhanced analysis of more data we can introduce predictive tools that can answer the question “Where will they be?” for a date and time in the future.

Previous toolsets have been restricted in the volume of data analyzed causing analyzed timeframes to be restricted to a number of days. This restriction has limited the forward planning capability of surveillance teams.

Applying a level of confidence selected by the analyst, planning, enhancing deployment of personnel
and their surveillance
technology. Field operatives have a higher degree of confidence for the future location of the target and as such can modify their tactical procedures for surveillance, equipment deployment/refresh and target interception.

  • Where will the target be?
  • Where will they have come from and go to after?
  • Which route will they have used?

Layered – social media, ANPR, phone, any geolocation data

With an array of data sets providing location attributes (lat:long) the creation of intelligence has become a more complex job.

Rapid import tools allow the analyst to import data sets from any source of location data. With the ability to create a standard import list and to import on the fly the addition of a new data set is no longer a time consuming headache.

Layering and de-layering the data sets allows for swift comparisons.

  • Mobile phone data
records – cell tower
data can be extracted and analysed to provide routes taken.
  • Smart travel card (eg Oyster) use – will enable a pattern of life to be created for public transport use.
  • Financial records – ATM cash withdrawals and payments
  • ANPR data – will place the target’s vehicle at a particular location and time.
  • Blue force data – can also place and locate known friendlies to determine compromise.
  • Others – any data set with time and location information.

Enhanced Pattern of Life context

Bringing these four new capabilities together enhances Pattern
of Life information and is delivered in a fraction of the time of traditional methods. The most immediate outcomes for this are:

  • Less time spent ‘crunching’ the data.
  • Data from a wide variety of sources can be easily tied together.
  • Answers to your most typical questions are generated automatically.
  • New answers are provided that would not have previously been possible