Intelligent Monitoring of Intelligence Data

在最近的恐怖暴行之后，情报专家评论说，未能发现恐怖活动与其说是因为缺乏数据，不如说是因为难以联系和解释现有的情报。监测和解释情报数据的自动化工具将为情报分析员提供一个有用的手段，让他们考虑可能发生的恐怖袭击的新情况，从而在制定和部署预防措施方面提供有用的协助。虽然传统的基于知识的系统，如基于规则和基于案例的推理机，在普通犯罪侦破方面有很好的应用，但它们的范围仅限于预见的情况或以前遇到的案件造成的情况。为了设计一个强大的监测系统，能够识别特定类型的恐怖活动的许多变化，该项目将采用基于模型的方法来生成情景。这类工作的主要潜力是它能够通过动态组合可重用的场景片段，从相对较小的知识库中自动构建给定类型的场景的许多变体。拟议的工作涉及一些全新的研究活动。特别是，由此产生的成分建模方法将是第一个能够同时处理多个场景的方法，第一个可以处理定义不明确的概念的方法，以及第一个结合定性偏好处理的方法。此外，应用这些方法来帮助监控情报数据本身也是非常新颖的。这项研究将为开发软件系统提供理论基础，这些软件系统将帮助反恐情报分析员尽可能广泛地考虑可能反映有组织恐怖活动的各种新兴情景。这项工作将有能力将看似不同和无关的情报数据联系起来，并将这些数据与逻辑推断和合理的可能情景联系起来。为该项目组建的多学科团队包括人工智能、反恐和情报研究方面的领先研究人员，为承担这项工作提供了非常强大的组合。

In the wake of recent terrorist atrocities, intelligence experts have commented that failures in detecting terrorist activities are not so much due to a lack of data, as they are due to difficulties in relating and interpreting the available intelligence. An automated tool for monitoring and interpreting intelligence data will provide a helpful means for intelligence analysts to consider emerging scenarios of plausible terrorist attacks, thereby offering useful assistance in devising and deploying preventive measures against such possibilities.While conventional knowledge-based systems, such as rule-based and case-based reasoners, have useful applications for common crime detection, their scope is restricted to either the situations foreseen or those resulting from previously encountered cases. To devise a robust monitoring system that is capable of identifying many variations on a given type of terrorist activity, this project will employ a model-based approach to scenario generation. The main potential of such work is its ability to construct automatically many variations of a given type of scenario from a relatively small knowledge base, by dynamically combining reusable scenario fragments. The proposed work involves a number of research activities that are entirely new. In particular, the resulting compositional modelling methods will be the first which are able to work with multiple scenarios at one time, the first which can handle ill-defined concepts, and the first to incorporate qualitative preference handling. Also, the application of these methods to aid in monitoring intelligence data is itself highly novel. This research will provide theoretical foundations to develop software systems that will aid anti-terrorist intelligence analysts in considering as widely as possible the range of emerging scenarios that may reflect organised terrorist activities. The work will have the ability of linking up seemingly distinct and unrelated intelligence data and associating such data with logically inferred and justified possible scenarios. The multi-disciplinary team assembled for this project involves leading researchers in artificial intelligence, and in counter-terrorism and intelligence studies, offering a very strong combination to undertake the work.