[目的/意义]大数据时代下的情报分析面临着分析方法复杂、技术难以理解、结果不易解释等方面的挑 战。可解释性技术是解决该类挑战、支持用户理解并信任情报分析决策结果的技术手段。[方法/过程]文章首先 介绍了情报分析活动的发展动向，并从情报资料搜集、情报加工、情报分析和决策服务四方面阐明情报分析流程 的解释必要性，接着阐述情报分析方法的解释类型，包括全局解释与局部解释相结合、以事后解释为主、群体智能 决策解释和以人为本的交互解释。[结果/结论]情报领域中情报分析方法的解释技术包括因果推断、特征重要性 分析、规则解释、知识推理和可视分析。情报分析解释中可用的评价方法有基于主观判断的定性分析、基于启发 式的半定量分析和基于指标评价的定量分析。最后，对不同领域情报分析主要的可解释性技术和未来研究应重 视的科学问题进行总结。

[Purpose/significance] Intelligence analysis in the era of big data faces many challenges such as complex analytical methods, difficult to understand technology, and difficult to interpret results. Interpretability technologies are technical means that enable users to understand and trust the decision-making results ofintelligence analysis. [Meth? od/process] The paper first introduces the development trend of intelligence analysis activities, and expounds the inter? pretation necessity of intelligence analysis process from four aspects:intelligence information acquisition, intelligence processing, intelligence analysis and decision-making services. Then, the interpretation types of intelligence analysis methodsare elaborated, including the combination of global interpretation and local interpretation, post-hoc interpreta? tion, swarm intelligence decision interpretation and people- oriented interactive interpretation. [Result/conclusion] The interpretation technologies of information analysis methods in the field of intelligence include causal inference, fea? ture importance analysis, rule interpretation, knowledge reasoning, and visual analysis. Commonly used intelligence analysis interpretation evaluation methods include qualitative analysis based on subjective judgment, semi-quantita? tive analysis based on heuristics, and quantitative analysis based on index evaluation. Finally, the main interpretable techniques of intelligence analysis in different fields and research issues that should be emphasized in the future devel? opment are summarized.