| Information imprecision and uncertainty exist in many real-world applications. It is especially true in the economic, military, telecommunications and other areas. Classical data models often suffer from their incapability of representing and manipulating imprecise and uncertain information. Traditional database based on binary logic can not express data of many ambiguous objects.For this reason, Zadeh’s fuzzy logic has been used to extend various data models since the early1980’s. The purpose of introducing fuzzy logic in database modeling is to enhance the classical models such that uncertain and imprecise information can be represented and manipulated. With the establishment of system of fuzzy set theory, people can use to describe the relationship between the number of fuzzy objects and fuzzy operations can be carried out.Since the objective reality world is a four-dimensional world, so each thing has its time dimension, along with the information flow of physical flow including the temporal information. Events and the potential relationship between them often contain the temporal information implicitly. However, many temporal information such as the end or start of the events are inaccurate in real life, it is often expressed as uncertain and ambiguous forms. And thus need management of fuzzy temporal data on demand.Database is widely used for data management related activities, such as data storage, data processing and information retrieval, and plays an important role in achieving fuzzy temporal data management exertion. The traditional database system process temporal datum only as a general attributes, which could not be handled and performed as fuzzy temporal information. Fuzzy database is able to handle fuzzy information, which is one of the important databases research. And provides an important way of management imprecise, incomplete and vague information, it not only has important theoretical value, but also has important application value. However, it should be noted that, although researches on imprecise and uncertain information and processing capabilities of fuzzy databases (including the fuzzy conceptual data model) have been made great achievements, there is no widely accepted for fuzzy temporal database model. And researches on fuzzy temporal database model are few. The theory of fuzzy temporal database system has not been formed yet, and issues of the model still require further investigation.Based on fuzzy logic and fuzzy sets theory, the fuzzy temporal interval were extended not only to express the precise relationship of fuzzy events, but also to describe the imprecise relations of fuzzy events. A fuzzy temporal database model is presented to process the fuzzy temporal information that the classical database cannot describe, enriching the temporal database query language. A kind of INF (first normal form) relational algebra available to describe fuzzy temporal information is established by defining a series of fuzzy temporal elements via analyzing the basic temporal database model because the complex data are not only temporal but also fuzzy. Based on the correlation of fuzzy query results and threshold, a-cut set of fuzzy query translation model was proposed on framework of fuzzy temporal query. Furthermore, based on the proposed fuzzy temporal model, this analysis characteristic of vector space on fuzzy temporal database, and introduces a formal mapping method of the multi-granularity fuzzy temporal functional dependency.Firstly, the description of fuzzy temporal that can effectively represent and deal with complex object attributes is the key to make the temporal database description of the objective world in accordance with actual abstract. Based on the theory of fuzzy sets, an expansion of Allens fuzzy temporal interval model for temporal interval is derived, and the fuzzy relation between temporal intervals is discussed in view of classical tense relations. Based on the analysis of measurement of fuzzy temporal information, extracting the fuzzy value corresponding to tuple discrete temporal interval, and these fuzzy values are mapped to discrete temporal interval, realizing the correspondence between temporal convex interval and fuzzy values at the fuzzy temporal database conception level. A fuzzy temporal mapping model of fuzzy set theory is established, and the semantic model that present in this paper realizes the mapping of fuzzy measure values to discrete temporal interval.Secondly, for the temporal and fuzzy characteristics of complex data, we define a series of fuzzy temporal relations elements and establish a1NF relational algebra to describe the fuzzy temporal information based on the basic model of the temporal database. Theoretically demonstrate the corresponding mapping of fuzzy temporal relational algebra and calculating model. Define a series of operation elements of fuzzy temporal relations, demonstrate the syntax and semantics of fuzzy temporal relational algebra, and set up a algebraic theory to describe the relationship between the fuzzy temporal information. A fuzzy temporal query method is proposed based on the constrained fuzzy temporal constraints, the query optimization and query evaluation of fuzzy temporal constraint is discussed. From horizontal and vertical aspect, the fuzzy temporal query processing and methods, algebraic optimization depended on the expression of fuzzy temporal predicate, and query translation for supported system optimization and evolution are described.Finally, characteristics of the multi-granularity of temporal type of semantic model based on fuzzy set are discussed. And the closed set of multi-granularity temporal type set, finite closure of attribute set, limited dependent base of attribute sets, dependency-based vector of attribute characteristics are defined. Algorithm of the determination of tuple similarity and so on, is proposed and proved. And then obtain the description of characters of fuzzy temporal vector space. |
PDF Full Text Request