不确定性人工智能数据的证据分析

Evidence Analysis of Uncertain Artificial Intelligence Data

由于人类在认识过程中表现出的智能和认知不可避免地伴随着不确定性,不确定性人工智能电子数据的可信性和证据分析的科学性问题也成为人们关注的重点。在司法实践中,需要通过区分基于有形物和基于语言(包括自然语言与人工语言)的证据属性来审查不确定性人工智能电子数据的可信性。同时,基于贝叶斯网络,实践中还需要在不确定性人工智能的信息点挖掘、模式识别以及证据表达方式三方面充分利用规则表示概念之间的关系,建立不确定性人工智能分析电子数据的认知模型,并用拓扑势形式化地表示电子数据信息中的规律性,这样才能确保不确定性人工智能电子数据证据分析过程的科学性。

Since the intelligence and cognition that the human beings show in the cognitive process are inevitably accompanied by uncertainty, the credibility of artificial intelligence electronic data with uncertainty and the scientific nature of evidence analysis also become a concern. In judicial practice, the credibility of artificial intelligence electronic data with uncertainty needs to be examined through distinguishing the evidence attributes between tangible electronic data and language(including natural languages and artificial languages) electronic data. At the same time, based on the Bayesian network, it is also necessary to make full use of the relationship between rule representation and concepts in three aspects of information point mining, pattern recognition and mode of evidence expression of artificial intelligence electronic data with uncertainty, to establish a cognitive model, and formally represent the regularity in electronic data information by using topological potential, so that we can ensure a scientific analysis process.