Research On Ontology Representation And Retrieval Of Temporal Database In The Emergency System

Time is an objective attribute of the nature world, all information has the corresponding temporal property. With the development of database and information technology, emergency systems have many new applications and requirements, and the temporal information processing is more and more urgent. The occurrence in emergency system has the characters of emergency, unexpected, time-varying etc. Emergency systems need to consider the temporal attributes of knowledge and data, enabling the system handle emergency cases correctly.In a typical temporal database (Bi-temporal database), the temporal information is recorded in time, the time attribute of real world information is reflected timely. Temporal data are a collection of all the data, it can provide data at instant and time interval, which is not likely for traditional database. Temporal databases provide data for comparing, analyzing and forecasting, so it can serve to management information systems, emergency decision systems, as well as other systems. Bi-temporal databases can store the changing status of databases and real world, so it can manage data effectively. Bi-temporal databases need a large amount of storage space and are hard to store, because of transaction time and valid time. It is necessary to change the temporal database's storage means and establish a temporal database presentation model, to solve the problem of temporal database's storage. The retrieval that based on ontology, can take full advantage of the structure description of data and relationship of concept, so it can realize intelligent retrieval. Now there is not a method for ontology presentation of bi-temporal database, because bi-temporal database is difficult to store.First, this thesis introduces essential knowledge of temporal databases and ontology presentation, and proposes a bi-temporal database presentation model and realizes the storage of bi-temporal databases, based on the bi-temporal conceptual data model. We then use the nested relational data model to give the formal description for ontology presentation of temporal data, and use ontology language to express the actual databases in emergency systems. Finally, we develop a semantic retrieval system, base on the above ontology. Through the analysis of the experiment results, show that semantic retrieval system can realize temporal database query. The semantic relations are comprehensive and clear, and it is in favor of emergency decision and knowledge reasoning in emergency systems.