Epistemic Programming System Based on Strong Relevant Logic

1.我们对科学发现过程及其自动化提出了一些重要的基本观察和假设。基于观察和假设，我们提出了科学发现认知过程的强关联逻辑模型.基于科学发现中认知过程的强关联逻辑模型，提出了一种新的程序范式--“认知程序设计”（Epistemic Programming），它以条件句为计算主体，以基本认知操作为计算基本操作，以认知过程为程序设计主体.指出了经典数理逻辑及其各种经典和非经典保守扩展不适合于自动定理发现的原因，并指出强相关逻辑是一个更有希望的候选者.我们开发了一个通用蕴涵演算的前向推理系统EnCal，它可以作为认知程序设计系统中的前向推理引擎。通过并行处理技术提高了EnCal的效率.提出了自主进化信息系统的概念和体系结构.我们提出了预期推理-反应系统的概念，并证明了在强相关逻辑中引入时态算子和相关公理模式及推理规则所得到的时态相关逻辑可以作为预期推理的基础.提出了一种新的知识获取方法：基于强关联逻辑的关联推理知识自动获取。

1. We proposed some significant fundamental observations and assumptions on scientific discovery processes and their automation. Based on the observations and assumptions, we proposed a strong relevant logic model of epistemic processes in scientific discovery.2. Based on our strong relevant logic model of epistemic processes in scientific discovery, we proposed a novel program paradigm, named 'Epistemic Programming,' which regards conditionals as the subject of computing, takes primary epistemic operations as basic operations of computing, and regards epistemic processes as the subject of programming.3. We pointed out why the classical mathematical logic and its various classical and non-classical conservative extensions are not suitable to automated theorem finding, and shows that strong relevant logic is a more hopeful candidate for the purpose.4. We developed a forward deduction system for general-purpose entailment calculus, named EnCal, which can serve as the forward reasoning engine in an epistemic programming system. We improved the efficiency of EnCal by parallel processing techniques.5. We proposed the notion of autonomous evolutionary information system and its architecture.6. We proposed the notion of anticipatory reasoning-reacting system and showed that temporal relevant logics, which are obtained by introducing temporal operators and related axiom schemata and inference rules into strong relevant logics, can be used to underlie anticipatory reasoning.7. We proposed a new approach to knowledge acquisition problem : automated knowledge acquisition by relevant reasoning based on strong relevant logic.