Research And Implementation Of Content Security Technology Based On Ontology

Content security, especially content security in web environment, has become a hot issue in modern society, being highly concerned by individuals, enterprises and governments all around the world. The beginning of this paper briefly discussed the current situation of content security and problems already faced, and then made a general introduction to Information Extraction and Content Filtering, which are two core technologies in content security field.Then the paper intently introduced ontology theory and semantic web technologies, illustrating the technical background of the research we have undertaken. Ontology is a methodology brought up in Knowledge Engineering. As a formal representation of knowledge, it is in some ways suitable for textual semantics description. Relating researches concerning the combination of ontology technologies and IE technologies have already made big progresses. The future web is semantic web. It embodies information with structure and semantics, purposed to construct a platform for information sharing and to make information machine process possible. The semantic web provides a better mechanism for content security management in one way, but brings challenges to old content security technologies in another.In chapter 4 the paper discussed the Content Analysis Ontology in detail. Content Analysis Ontology was designed for a content security related project that we have made. It was purposed to give a domain knowledge representation for content security, and based on that, to explore new approached for advanced content security techniques. In this part we explained the architecture, content and functions of the Content Analysis Ontology, and explained relevant modules that were designed for ontology management and maintenance. Charter 5 also depicted the reasoning system that was highly interrelated to CAO, including theoretical foundation, reasoning tool, system architecture, and modules designed for the implementation of the reasoning system.The paper introduced three content security technologies, which are all based on ontology: the Ontology-based Information Extraction: the Ontology-based Textual Tendency Analysis and the Ontology-based Content Classification. All of the three technologies are within IE and Content Filtering field.Chapter 6 was about the Ontology-based Information Extraction system. As the implementation of extraction system was not the main work of the author, we only emphasized on how the ontology serves the IE system other than the IE system itself. The introduction of the extraction system included the architecture, techniques and implementation tools etc, and the Triple Extraction was also explained here for it is particularly meaningful to the whole system; in the rest of this chapter the paper discussed how the ontology supports the IE system, including the design and implementation for the lexicon base and for the ontology parsing module.Chapter 7 focused on the Ontology-based Textual Tendency Analysis. This part mainly expatiated the idea of the approach, described the algorithm of the analysis method, explained the tendency extraction system-including system structure and implementation; and in the end of this chapter the paper made a simple performance analysis on this approach according to the experiment data. Chapter 8 explained the Ontology-based Content Classification. The paper firstly made a comparison between traditional text classification techniques and the technique introduced in this paper, and then explained the principle of the content auto-classification, which actually takes advantage of the reasoning based on ontology; the following part of this chapter explained the classification module-including the design and the implementation, and the core algorithm of the module were described also; in the last part of this chapter the paper introduced some optimization approaches for the classification system, correlated performance analysis was also given in detail.