The revolution in information technology has meant that the world is more interconnected than ever before. We have easy access to a range of devices that are able to capture text, images, audio and video of events or incidents and share them across platforms such as Facebook, Twitter, YouTube, Periscope, Telegram as well as Intelligence Fusion.
As the rapidly evolving digital landscape fuses with, and consequently changes, the security industry, we’re capitalising on a society that views the world through their phone in order to enhance our threat intelligence provision.
What is Crowdsourcing Intelligence?
Crowdsourcing for us is the ability to not only engage a network of active, motivated participants who are providing us with incident data, but also tapping into their collective knowledge which can provide us with local context. The kind of granular context that other intelligence providers don’t have access to.
But it’s not just intelligence companies who can benefit from crowdsourcing data. There are many sectors who could improve the quality of their information by simply engaging their workforce or community.
Crowdsourcing is the action of tapping into the collective intelligence of the public at large.
Collective intelligence is shared or group intelligence that emerges from the collaboration and collective efforts of many individuals and appears in consensus decision making.
Crowdvoting is when a website gathers a large group’s opinions and judgements on a certain topic.
Crowdsolving is a collaborative way of solving a problem using many people. It is a type of crowdsourcing with focus on complex and intellectually demanding problems requiring considerable effort, and quality/uniqueness of contribution.
Crowdsearching is the process of crowdsourcing the location of things, be they pets, items or people.
Real World Value
I became convinced that crowdsourcing was a highly powerful tool when I was in Basra in 2011. At the time I was connected to other intelligence analysts and security managers in the province via a Skype chat group. If I received initial reports of an incident, for example an explosion in the city, I would report and share those details with others in our group chat. Throughout the day, as people began to hear other snippets of information, we would update each other and together, achieve a consensus as to what happened in the incident.
It was then when I knew I wanted to create a platform that was built specifically for the purpose of crowdsourcing information globally as well as disseminating large quantities of complex data in a user friendly environment.
The Evolution of Intelligence
Private intelligence is a relatively new service in the private sector. Historically, it was only large corporates who could afford this type of support, however, with advances in technology this service is now much more affordable.
Private intelligence has evolved from the use of internal teams of analysts to gather, process and disseminate intelligence in report format to technological solutions, such as data mining and intelligence platforms.
We believe the next evolution in intelligence is the use of global crowdsourced networks. Fused with our current methods of data mining and an internal team, we can create a far larger, more consistent data set.
Clients will also benefit from greater insights as well as deeper context and an increase in speed. Every peice of information is moderated and processed into actionable intelligence by our internal analysts. Becasue our expert team are evaluating large quantities of data, they can see both the tactical and strategic picture in their areas of intelligence responsibility, improving the quality of their assessment.
Reciprocity is Key
In order to create a motivated crowdsourced network of users there has to be a transaction of value between participants. We want users to provide us with incident data that we may be missing and in return we’ll provide them with free access to our platform.
If users want unrestricted access to our data or to use our premium features, such as the ability to filter the map, or view analysis and statistics, they can do this by submitting incidents to us for moderation. For the user they’re getting access to one of the foremost intelligence data sets globally, and in return, we’re receiving incident data that we’re yet to identify. It helps with the global breadth of our reporting but also the depth of our incident data.
As an example, if a high profile attack occurs, those free users can provide us with incident details at the time of the attack and unlock credits for our system. In the following days, weeks and months as more details regarding the attack are released, or the attacker goes on trial, users can continue to submit details to unlock more data and new features.
This is always supplemented by our internal team and data mining algorithms, but crowdsourcing can provide additional capacity as well as foreign language coverage and local context – all of which are invaluable to our organisation.
‘But how do you guarantee the quality of the information?’
We ensure the integrity of our data through training and best practices in intelligence doctrine, as well as technical means.
We have our own internal intelligence training programme, which all of our analytical team must go through. The training includes The Intelligence Cycle, with emphasis on the processing phase which includes evaluation of sources.
From a technical perpective, we’ve built a moderation area within Intelligence Fusion 2.0. Every incident that is reported by our crowdsourcing intelligence network (and that identified by our datamining technology) is sent to be verified and approved before it goes live on our system. What’s more is that once our analysts have reviewed an incident, it can be forwarded to another analyst for further checks if required.
Every incident on our platform must also have at least one source of information to ensure accuracy, which we upload alongside the incidents for our clients to view. By providing our clients with our references they then also have the due diligence to review where our incident data came from.
Finally, during moderation every source is graded in terms of Reliability and Credibility. Let’s say we receive a tweet regarding an incident but this is the first time we’ve seen any information from this source, we’d initially grade them as an F for Reliability. That means that their reliability cannot be judged because we currently have no other data to make a comparison.
However, if we receive more information from them at a later date, we would adjust that Reliability rating as appropriate.
|D||Not Usually Reliable|
|F||Reliability Cannot Be Judged|
We also grade the Credibility of the information from 1 – 6. 1 means that the information was confirmed by other sources whereas 6 means that the truth cannot be judged.
|1||Confirmed By Other Sources|
|6||Truth Cannot Be Judged|
By using these scales we can actively track how reliable sources are, but also the credibility of the information they provide. Whilst the majority of incidents we report on our platform will likely come from news media or social media, as we move to reporting directly from the source through crowdsourcing, this source grading capability will help us to ensure the integrity of data.
As a further example, if someone witnesses a crime and reports that incident to Intelligence Fusion as the primary source, initially, we’d likely grade the source as F-3 (Reliability Cannot Be Judged – Possibly True). We would then look for other sources to confirm the incident before we approve it in moderation and disseminate it via our platform.
As our trust in the source grows, we may get to a point where they have proved their reliability and we will upload the incident as single source reporting but from a trusted source.
Crowdsourcing and High Profile Attacks
When a high profile attack occurs, Intelligence Fusion will internally swarm that incident. This means that the first person to identify the incident sends out a company wide message and reports the initial details to our clients via the platform.
As a team, we’ll gather as much information surrounding that incident in as short a period of time as possible. The idea being that many hands make light work and we can quickly get to grips with what happened, as well as providing additional context and the ‘So What?’ analysis.
Through crowdsourcing, we want to take this process global. When a terrorist attack occurs threads begin to appear on Reddit and the community will provide links and details on what they’ve discovered. However, Reddit is not a platform that has been built for the collation of this type of data.
When these incidents take place, there’s a significant duplication of effort in both the public and private sector. Companies, organisations, military, police and security services are all collecting the same information about the same incident. Duplication of effort can be considerably reduced with Intelligence Fusion.
By creating a central node around which all the above organisations can gather, this will allow for the rapid collation and processing of information regarding the incident, as well as ensuring that accurate information is cascaded as quickly as possible.
Crowdsourcing for the Common Good
Europol have an initiative called ‘Stop Child Abuse – Trace An Object’. On the website it states;
‘The most innocent clues can sometimes help crack a case. The objects are all taken from the background of an image with sexually explicit material involving minors. For all images below, every other investigative avenue has already been examined. Therefore we are requesting your assistance in identifying the origin of some of these objects. We are convinced that more eyes will lead to more leads and will ultimately help to save these children.’
This is an example of an initiative that we’d like to support through our own approach to crowdsourcing intelligence. With a global network of users, we can tap into their collective knowledge and assist in the identification some of those objects or locations, consequently helping Europol identify offenders and prevent child abuse.
Waze has reportedly over 100 million users and is described as the world’s largest community based traffic navigation app. They state on their website;
‘Imagine millions of drivers out on the roads, working together towards a common goal: to outsmart traffic and get everyone the best route to work and back, every day.’
The idea behind this is, the more people who provide data the more accurate it will be.
The same principle can be applied to intelligence. The more people providing data regarding global security threats, the more accurate the data will be.