Home Editor's Picks How Machine Learning Is Changing Intelligence Collection

How Machine Learning Is Changing Intelligence Collection

How Machine Learning Is Changing Intelligence Collection

This article was posted originally to InPublicSafety.

By Eric W. AdamsAlumni, Master of Intelligence Studies at American Military University

There is an extraordinary amount of data being generated around the world on a daily basis. The task of collecting relevant information, organizing it, and piecing it together in a way that tells a story seems like an overwhelming and nearly impossible task. Yet, this is the monumental task of intelligence analysts.

Get started on your cybersecurity degree at American Military University.

Thankfully, incredible technological advancements, including machine learning (ML) and artificial intelligence (AI), are assisting analysts in their efforts to collect and categorize massive amounts of data. Technology is evolving at a rapid pace, and analysts must always be learning how to apply machine-learning technology to help them better understand and solve complex problems.

[Related: GIS Technology Aids in Effective Response to California Wildfires]

When I was a student earning a master’s degree in Intelligence Studies at American Military University (AMU), I learned about these technologies and how critical they were for the intelligence profession. As an intelligence professional today, I was able to take what I learned in the classroom and apply it directly to my job.

How Machine Learning Aids in Data Collection

One of my primary job responsibilities is to collect many types of data such as spatial data, which also includes satellite imagery. I use these data sources to look for objects of interest. However, it can take an analyst a very long time to visually and manually scan individual images for specific objects. Using ML and AI can automate object detection and enhance the efficiency at which intelligence is derived.

In order to use these technologies, an analyst must first start by “training” data, which includes a step called “labeling.” In this step, analysts have pre-captured image examples about what a particular object looks like.

For example, say the objects of interest are buildings. The analyst finds several example images containing building types. The first step is to capture the many different attributes of a building and add it into the labeling software (see image below). Such attributes include types of buildings such as residential, military, or commercial. In the training images, the analyst finds buildings and “annotates” them using rectangles or polylines, and then chooses the classification (type) of the object based on the attributes. The image is now labeled and included in the software.

Characteristics of features are captured and placed in the left column for analysts to attribute features.
(Image used with permission by LabelBox).

This step is repeated over and over for as many images are needed (usually in the hundreds or thousands). The more training data input into the system, the better the output results will be. These “labels” are then exported as a JavaScript Object Notation (JSON) file, which is then applied to the development of a machine learning model. These models are then used to infer objects within images—a term referred to as inferencing.

Then, an analyst can simply connect a database of images that haven’t been reviewed yet and the computer will scan each image, detect the object (in this case, buildings), auto-classify them, and return the found objects (flagged in the image) to an analyst for review. Using this technology can provide immense enhancements in processes, workflows, and quality output.

AI and ML Improve Intelligence Collection, but Human Input Still Needed

Even with this technology at the fingertips of an analyst, it important to remember that intelligence remains an art. Upon receiving these detected objects, the analyst must still infer the meaning of the active presence of the objects and the changes in these objects’ positions over time.

For example, an analyst might be focused on an intelligence problem related to piracy. As a result, he or she is interested in the daily activities within a given harbor and only interested in the activity of commercial ships. Using these advanced applications in object detection, an analyst could track when and where commercial ships are at given times or certain days to determine if there are patterns in pirate attacks. Another application might be to determine the level of shipping activity in that harbor or port. Using advanced object detection can help analysts identify such specific activity and collect related information so they are able to better understand a complex situation.

[Related: Why Operational Dashboards Are Vital for Emergency Management]

Conducting Assessments Requires Critical Thinking Skills

As an analyst, one must be prepared (and willing) to challenge him or herself to conduct assessments made about a particular event. However, the “why” is not always clear. Understanding why events happen is often complex and individuals are constantly at risk of offering biased opinions. When analysts are attempting to apply a reason for an event, they have to be careful to use the information they have and not apply personal aspects or make assumptions about activities. This can be incredibly difficult and where training from both professional and academic experiences kicks in.

The courses I completed at AMU focused on enhancing critical thinking skills and intellectual approaches to decision making. A computer cannot do the thinking for you.

In my courses, for example, I learned how to employ a set of assessment techniques based on the publication “Structured Analytic Techniques for Intelligence Analysis.” I also learned to implement strategies of thinking from Richards J. Heuer’s “Psychology of Intelligence Analysis.” In addition, I learned to employ the technique of devil’s advocacy and the analysis of competing during my coursework. These techniques force critical evaluation of facts, assumptions, and perspectives that contribute to bias and dramatically improve decision making.

Combining these analytical techniques with advanced technologies can lead to a more comprehensive and higher quality intelligence product, derived more efficiently and likely more quickly. However, to make such improvements in intelligence, analysts must always be willing to learn new assessment skills as well as understand how to apply new advanced technologies.

About the AuthorEric W. Adams graduated from American Military University with a master’s degree in Intelligence Studies and is currently in the application process for the AMU program of Doctor of Strategic Intelligence. Additionally, he is currently active in a graduate program with North Carolina State University’s College of Natural Resources, Geospatial Information Sciences and Technology (MGSIT). Eric is an Army veteran with more than 21 years of geospatial intelligence experience and currently works as a government contractor for the National Geospatial-Intelligence Agency (NGA) in Springfield, VA. His primary focus is developing capabilities to provide advances in Machine Learning and Object Detection to the customer. To contact him, email IPSauthor@apus.edu. For more articles featuring insight from industry experts, subscribe to In Public Safety’s bi-monthly newsletter.