面向关联推理的智慧情报感知:内涵、组织与路径

Intelligence Perception for Association and Reasoning:Connotation,Organization and Path

[目的/意义]海量异构的开源数据已逐渐成为科学研究的关键生产要素,如何关联多源信息,实现情报的自动推理和感知,是迈向下一代智慧情报系统的必要途径。[方法/过程]文章探索性地提出了智慧情报感知框架,围绕情报关联和情报推理两个核心组件,论述其内涵、组织要素和支持路径。认为智慧情报感知的内涵是依托海量异构的开源数据,利用情报方法实现情报任务的主动观察和主动发现的过程。在组织方面,分别介绍情报关联组件,即情报评估,抽取,融合;情报推理组件,即情报演进刻画,溯因,预测。在路径方面,分别阐述涉及关联与推理的内部支持路径和涉及人员、方法、任务的外部支持路径,并讨论各组织要素之间的运行逻辑和协同关系。[结果/结论]智慧情报感知旨在拓展现有情报感知体系的方法范畴,将因果推断、事件溯因、轨迹刻画等前沿技术、方法引入感知研究,将数据感知和机器感知深化至智慧感知。

[Purpose/significance]Massive heterogeneous open source data has gradually become a key production factor of scientific research.How to associate multi-source information and realize automatic reasoning and perception of intelligence is a necessary way to move towards the next generation of intelligent intelligence system.[Method/process]This paper tentatively puts forward the framework of intelligent intelligence perception,and discusses its connotation,organizational elements and support path around the two core components of intelligence correlation and intelligence reasoning.We points out that the connotation of intelligent intelligence perception is a process of active observation and active discovery of intelligence tasks by using intelligence methods based on massive heterogeneous open source data.For the aspect of organization,we introduces the intelligence related components,namely intelligence evaluation,extraction and fusion.Intelligence reasoning component,that is,intelligence evolution,characterization,abduction and prediction.In terms of path,this paper expounds the internal support path involving relevance and reasoning and the external support path involving personnel,methods and tasks,and discusses the operation logic and collaborative relationship between various organizational elements.[Result/conclusion]Intelligence perception aims to expand the method category of the existing intelligence perception system,introduce cutting-edge technical methods such as causal inference,event abduction and trajectory characterization into perception research,and deepen data perception and machine perception to intelligent perception,which can promote the integrated development of China’s intelligence work mode from active perception observation and scene task reasoning.It provides theoretical and methodological support for the correlation,induction,reasoning and decision-making of various intelligence tasks,and provides a novel research perspective for industrial intelligence,competitive intelligence,security intelligence and other fields.