Research On The Structure And Catalog Of Concept Element Space

As the basic unit and form of logic thinking, conception is our cognition of theessence of things. The research on the structure of the system formed by concepts andits evolution mechanism is a new field in knowledge representation and treatment.Aiming at the formalization expression of concept, this paper puts forward aconcept-element space as the basic expression frame of concept system to describe theformalized definition of the connotation, denotation of knowledge and relationshipbetween them and to deal with the information dissemination of the inner informationof the same level as well as between different levels. In order to expressconcept-element specifically, we describe its semantic form as granule structure(denotation, connotation and context relationship). In terms of its inner structure, anidea is proposed as follows: concept-element space is the community composed ofsome bottom elements about the knowledge shared by all. The constructed conceptgranule can be used as sensor and filter to receive knowledge from outside for thewhole system to compress the knowledge within the boundary as kernel knowledgeand dispel irrelevant knowledge elements out of the boundary of concept granule.The overall objective of this paper is: the relevant meaning of elements withinthe system can be represented in a formalized way by using semiotic structures inorder to study the relevance and coordination between elements of concept-elementspace as well as the special structure and functions formed by these elements to builda static model; parts and whole of the system are linked together organically withintime dimension to study the time unipolarity featured by concept-element space, theunchangeability and dynamics of the evolution system structure, the exchangeabilitywith external environment so as to construct a dynamic evolution model for thedissemination essence of concept-element space to forecast and control its behavior.Our innovation and research results are as follows:1) Based on the isomorphism relationship between FCA, L-fuzzy set and roughset respectively, a formalized definition of granule structure is constructed to make a formalized description of the knowledge body of concept system. Simultaneously, themeasurement about relevance and relevant degree within and outside the conceptelements is given to reflect the similarity and coincidence between the objects withinand between concept elements. Moreover, based on the Galois connection, the generalform of mutually relevant concept aggregation is presented as concept granule and theconnotation, denotation and structure of concept granule are investigated in detail.The relevance distance is set and proved as a measurement, revealing that the concept-element space induced by relevance distance is such a space that can be measured.Therefore, the concept-element space is attached with topological structure. Inaddition, the cohesion and the cohesion in a broad sense of concept-element space isalso studied to identify the concrete form of interaction between concept elements ofdifferent levels.2) A causal system called environment space containing regulation and rules ofintegration function between knowledge bodies is put forward. The rules contained inenvironment space are investigated in the frame of FCA, L-fuzzy set and rough setrespectively and the completeness of environment space is proved within differentframes. In the environment space constructed by rough set, causal relationship isabstracted as relevance relationship between different elements to represent the fuzzycausality of different phenomena. By defining the character circle and its inferentialdegree, the most reasonable inference of the final causality in different relevant formsis given. Then association logic is proposed based on semantics, different withtraditional logic. The semantic inference of association logic is concluded as a processof finding convergence path of the highest inferential degree from manytranscendental phenomena in the frame of concept-element space. The inferenceresults obtained should not be expressed completely as the absolute true value of0or1, and thus all kinds of uncertainty caused by broken logic of the law of excludedmiddle can be dealt with to a certain degree.3) Based on the frame of concept-element space, the category of time-state andtime sequences are represented in a formalized way and phenomenon element andphenomenon space are put forward. The phenomenon element of time-sequencematter can be interpreted as the fundamental elements constructing knowledge. By introducing time sequence which tends to sequence relationship, the diachronism ofphenomenon is investigated and the isomorphic mapping is made between thesequence structure formed by time sequence matter and concept lattice structure tomake the movement of phenomenon visualized in time. The family similarity ofphenomenon evolution is put forward, and on that basis, the unchangeability of themovement of phenomenon in time is investigated in depth. Thus, a more completedynamic knowledge representation model can be constructed semantically.In this study, by means of formalized representation of the bottom elements ofconcept system and by defining time-sequence relationship, the concept system isdescribed as a structure composed of construction process, i.e., the structure of systemis attached with timeliness to lay a base for studying the openness of concept system,information flow and knowledge evolution and provide a general frame to buildmodels for the ontology movement of the concept system in real world. Moreover,association logic and semantic inference can simulate the dynamic process ofobtaining and studying knowledge by human beings and the nonlinear process ofinterpreting, reasoning out all kinds of phenomena in the objective world by using theexisting knowledge system and solving related problems.