GraphAware - Transforming policing with graph-based intelligence analysis

Petr Matuska, Sales & Sales Engineering Lead, GraphAware

Western Australia Police Force’s adoption of Neo4j and the GraphAware Hume graph analytics platform marks a significant advancement in data-driven policing. Facing the challenges of growing volumes of valuable data scattered in disconnected silos, the organisation successfully implemented Neo4j database and Hume, consolidating data from various sources into a dynamic knowledge graph. The result was a connected view of intelligence, making it easier for analysts to solve crime faster. The partnership between Neo4j and GraphAware in this project demonstrates the transformative impact of graph technology on law enforcement’s ability to leverage growing volumes of valuable data to prevent crime and protect communities.

1. 1© 2024 All Rights Reserved | GraphAware
Transforming policing with
graph-based intelligence
analysis
Petr Matuska

2. 2
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Petr
Matuska

3. 3
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
About Us
GraphAware has been helping Neo4j
customers achieve successfor over 10
years.
Since 2013, we've been pioneering the adoption of graph
technology across the globe.
We’ve written books and articles on how to apply graphs in law
enforcement, finance, cybersecurity, pharma and public sector.
We distilled our knowledge, experience and skills into Hume.

4. 4
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
●What is a graph database & why to use it in policing?
●What is Hume?
●How graphs solve policing data challenges

5. 5
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
●What is a graph database & why to use it in policing?
●What is Hume?
●How graphs solve policing data challenges

6. 6
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
PERSONEVENT LOCATION
Relationships can have
properties (name/value pairs)
:INVOLVED
Date:2008-01-01
{
Name:Amy Peters
Date_of_birth:1984-03-01
{
Nodes can have properties
(name/value pairs)
Relationships can have
properties (name/value pairs)
{
:LOCATED_IN
{
Nodes can have properties
(name/value pairs)
Description:Homicide Case
Date:2008-01-01
employee_count:1546

7. 7
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
ChartvsGraph
Chart Graph

8. 8
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
ChartvsGraph
Chart/TableGraph

9. 9
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
ChartvsGraph
Chart/TableGraph

10. 10
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
ChartvsGraph
Chart/TableGraph
Relational
Database
Graph
Database

11. 11
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
The world isconnected
and should be modelled as a graph.

12. 12
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Schema

13. 13
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Consider the most common
intelligence analyses
They are graph based questions.
●Show me how person X and person Y are interconnected.
●Show me how this gang is sourcing their funds
●Show me the communication events between two individuals
●Who are the most influential parties within a criminal organisation?

14. 14
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
But wait.Can’t this be done with existing tools?

15. 15
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
But wait.Can’t this be done with existing tools?
Only if you are willing to accept some significant limitations
and these limitations are hindering you more and more with every
new data point added.

16. 16
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
But wait.Can’t this be done with existing tools?
●Tabular tools will not be able to efficiently return complex queries
●They will not be able to enable complex pathfinding and graph
algorithms needed
●They will be slow and provide a poor use experience
●They will not support integration of new data sources easily

17. 17
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Finding Truth in Data

18. 18
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
●What is a graph database & why to use it in policing?
●What is Hume?
●How graphs solve policing data challenges

19. 19
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
What isHume?
Hume is a mission-critical graph
analytics solution that puts the
power of Neo4j native graph
performance in the hands of
intelligence analysts.
Hume provides a simple to use
interface that allows to build, search,
visualise and analyse graphs.

20. 20
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
EmbracingConnected Data
Hume is fully graph based up and down the stack, making analysis more efficientand effective.
Advantages of Graph Data
●Easy to Integrate
●Flexible with respect to Schema
●Intuitive and Fast to Query (especially for
Complex Pattern Matching)
●Understandable and Explainable
●Ready for DS and ML
Business Outcomes
●Single View of Intelligence
●Rapid Capability Development
●Deeper Intelligence Analysis & complex Alerting
●Productivity and Collaboration
●Advanced Analytics and Insights

21. 21
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
●What is a graph database & why to use it in policing?
●What is Hume?
●How graphs solve policing data challenges

22. 22
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Challenges theintelligence teams face
Siloed data sources
There is no central store of data, or
curated views that have been put
together.
Siloed analytical tooling
There is no single tool used by
everyone for everything. Makes
collaboration difficult
Relational stores for
network analysis
Intelligence analysis requires
analysis at depth. This is difficult to
do with relational stores.
Manual analysis
Along the entire process is manual,
time consuming steps.
1
2
3
4

23. 23
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Siloed data sources
Disconnected analysis
Analysts search through dozens of
data stores to find the information
they are looking for.
None of these data stores are
connected to one another.

24. 24
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Intelligence holdings are ingested into
a graph,and made available in Hume.
Creating a single view of intelligence
for your analysts.
We make them Searchableand Connected.
Siloed data sources
→ single view of
intelligence

25. 25
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Siloed analytical
capabilities
Analysis capabilities are often
conducted on multiple distinct
platforms leading to fragmented
analysis findings.
Geographic, network and temporal
analyses are often all done
separately.

26. 26
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
One canvas
Many analysis capabilities
On a single canvas understand the
when and where of crime, and
detect emerging trends within
criminal networks of interest.

27. 27
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Using tables
to analyse criminal
networks
In truth, intelligence analysts have
been using graphs for decades.
But they have been doing this
without graph native tooling

28. 28
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Discover unknownunknowns
across multi-hop queries via
pattern matching and path
findingin milliseconds.
Using tables
→ pattern matching

29. 29
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Manual
analysis
All of these challenges ultimately
mean that intelligence analysis has
several manual steps.
Analysts are often responsible for
data cleaning, extraction, loading,
and the entire lifecycle of data
analysis.

30. 30
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Use Actions
to streamline analysis and
data access
Actions allow users to run complex
queries on their entities of interest
with the click of a button.

31. 31
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Advanced Expand
Visual query builder
Enables end users to query build
queries without having to learn how
to code.
Simplifying complex query creation
Seamlessly navigate intricate datasets, redefine
initial data discovery, and embrace a no-code
approach. It streamlines graph analysis,
empowering users to efficiently explore complex
data structures.

32. 32
GraphAware.
|
Hume
© 2024 All Rights Reserved | GraphAware
Thank You
Mission Critical Graph Analytics