Hard and soft information fusion to aid situation understanding

这个项目的目的是设计和开发算法和方法，通过使用机器学习、模式识别和其他新的非结构化数据挖掘方法，从社交媒体和其他文档(软数据)中识别和提取与特定领域的情况理解相关的信息。这个项目利用了Patera教授的博士后M.-O.St-Hilaire的初步研究，他后来加入了Ooda Technologies Inc.(Ooda)，以及Patera教授和他的团队在李群理论中的信息分析方面的广泛研究，以开发管理、分析和从软数据数据中提取知识的方法，并通过将这些信息与从物理传感器(硬数据)获得的其他信息/观察相融合来评估它们对情况理解的贡献。李群理论方法可以在计算速度、内存和硬件要求方面帮助提高多维数字数据处理的性能。结合工业合作伙伴OODA在开放源代码情报(OSINT)信息处理方面的知识和专业知识以及使用其产品的可能性，这种硬信息和软信息融合的新方法可以帮助设计新的工具，以解决信息的可用性和质量问题，并提供必要的背景以建立态势感知(SA)和理解，以帮助在任何领域以自动化方式进行决策。科学贡献将在海洋使用案例中进行演示和评估，这些案例适用于可能无法获得常用传感器数据的情况，或可能使用海洋环境监测系统数据作为背景的情况。在加拿大北极，OSINT数据可以帮助克服自动识别系统(AIS)和合成孔径雷达(SAR)传感器的挑战(例如信息质量、身份模糊、检测概率、刷新率和延迟)，这可能会给SA和加拿大北极地区的理解带来潜在的好处。蒙特利尔大学的科学小组将有机会直接为最现代的信息融合趋势作出贡献，并维护加拿大的主权。

The aim of this project to design and develop algorithms and methods to identify and extract information from social media and other documents (soft data), which are pertinent to situation understanding in a specific domain by using techniques such as machine learning, pattern recognition and other novel unstructured data mining approaches. This project leverages the initial study by Prof. Patera's Postdoc, M.-O. St-Hilaire, who later joined OODA Technologies Inc. (OODA) as well as Prof. Patera's and his team's extensive research in information analysis in the theory of Lie groups to develop approaches for management, analysis and extraction of knowledge from soft data data and assess their contribution into situation understanding, through fusion of this information with other information/observations obtained from physical sensors (hard data). The Lie group theoretical approach can help improve multi-dimensional digital data processing performance in terms of computational speed, and memory and hardware requirements. Together with knowledge and expertise of the industrial partner OODA in Open Source Intelligence (OSINT) information processing and the possibility to use their products, this novel approach to hard and soft information fusion can help design novel tools for addressing availability and quality of information as well as bringing the necessary context to establish situation awareness (SA) and understanding to aid decision making in an automated way in any domain. The scientific contributions will be demonstrated and evaluated on the maritime use case for scenarios where commonly used sensor data may not be available or where OSINT data may be used as context. This could bring potential benefit to SA and understanding in the Canadian Arctic where OSINT data could help overcome Automated Identification System (AIS) and Synthetic Aperture Radar (SAR) sensor challenges (e.g.information quality, identity ambiguity, probability of detection, refresh rate and latency). The scientific team at the University of Montreal will have the opportunity to contribute directly to the most modern trend of information fusion as well as to maintaining Canadian sovereignty.