Staying Four Steps Ahead: Understanding and Predicting the Behaviour of Adversaries
1. Challenge Statement
The Department of National Defence and the Canadian Armed Forces (DND/CAF) are looking for innovative solutions to support Activity-Based Intelligence in order to assist analysts to understand and predict the behaviour and movement of entities in real time.
2. Background and Context
Activity-Based Intelligence (ABI) is a methodological framework for analyzing data sources. It is used to discover correlations, resolve unknown-unknowns, develop knowledge, drive data collection and make inferences based on having access to large data sets and large sets of observations originating from different but complementary data sources. Given the volume of data available, analysts can spend excessive time searching for data. In order to keep a step ahead of one’s adversaries, speed and accuracy are vitally important. New data driven automated tools, techniques and approaches are needed to supplement (not supplant) the work of Analysts by supporting their ability to filter data, automatically detect patterns, and test hypotheses.
There are several steps in the process. Before data can be aggregated, managed and exploited, it has to extract the geospatial and temporal information. All actions take place at a time and location in physical space. Algorithms are needed to extract spatiotemporal information from data sources when it is available and to rapidly and accurately infer missing information. The next step involves data fusion. Multiple and simultaneous sources of data are often needed to describe the activities of entities across time and space. Data must be integrated in order to establish context and support advanced spatiotemporal analysis. The goal is to build models of potential outcomes, anticipate what may happen and why in order to support decisions, such as when to forestall an anticipated event or action of an adversary.
3. Desired Outcomes
Develop and apply new and innovative tools, techniques and approaches to support real or nearly real-time ABI. Innovators must choose a problem that is deemed relevant to defence and security for which archival and real-time data sets can be accessed, aggregated, managed and exploited by the innovator.
The overall objective of the Challenge is to detect activities of interest, discover new activities of interest and to anticipate events or actions, by minimally aggregating geospatial and time information.
The desired outcomes of the challenge include, but are not limited, to the following:
- Ability to derive timely, predictive insights from the data set with respect to the problem;
- Ability to detect patterns and infer activities of interest;
- Means of rapidly pulling existing geospatial and date/time stamped data from all sources when it is available as part of the meta-data, and rapidly inferring data that is missing;
- Ways of simplifying the task of filtering massive amounts of data from multiple sources;
- Ability to combine and aggregate a minimum of three (3) different data sources and formats;
- Ability to identify and incorporate at least three (3) additional variables of interest for analysis;
4. Supplementary Information
Innovators must identify a real-world problem of interest that is relevant to defence and security for which large datasets can be readily accessed. No further information will be provided to applicants by DND/CAF.
Innovators are asked to submit solution proposals involving data that will maximize their ability to demonstrate their capabilities.
The following are limited examples of problem areas that innovators might choose to focus on:
- Sovereignty concerns such as illegal fishing, smuggling, migrant ships, Arctic incursions;
- Traffic accident prediction;
- Natural disasters;
- Predictors of political instability and political violence, etc.
Report a problem or mistake on this page
- Date modified: