The Research For Construction And Analysis Of Movie Domain Ontology

There are three types of the technology of the information Retrieval:Full Text Retrieval, Data Retrieval, and Knowledge Retrieval.The Full Text Retrieval and Data Retrieval are directly based on keywords retrieval technology, which not only makes the computer can’t correctly understand the semantic information, but also has serious query omission and error checking, at the same time, it can’t meet the needs of users in the semantic and knowledge level, and can’t be precise retrieval and personalized recommendation. Knowledge Retrieval has a better guarantee the precision and recall recall for its emphasis is on knowledge and semantic matching. The researchers proposed the concept of "Ontology" in order to ensure the accuracy and retrieve personalized recommendations, but also enable the computer to have the intelligence. Using of ontology technology to describe the semantic information, not only the nature of knowledge and information that can be tapped, but also can greatly improve recall and precision retrieval, which is truly personalized recommendations.Movie is now in the power of exponential growth as a social information dissemination, an important resource for the entertainment of public life. How to effectively organize and manage vast amounts of video resources, how to efficiently meet the users recommended psychological needs and personalized to the user’s work, are the primary problem for mobile network service providers; as well as how to quickly and accurately find what they want is the psychological tendency of Internet users for mobile network users.This article will provide an effective solution based on the above issues and combined with the current hot topic of the Semantic Web-Ontology technology for the description and management of video resources, recommendations and retrieval of video works. The main contents are as follows:1. Determine the classification structure information Filmography resources. Currently the major video sites in the film and television works largely based nature of the site and their positioning for targeted division, which did not do popularization and standardization. This thesis will sort out a suitable public demand and easy to learn from each site film classification framework for present mainstream video site Filmography classification research and analysis.2. Building film domain ontology. There does not have a full field of ontology construction methods and standards. The thesis proposes a fitness method with an in-depth study of the major video sites in the classification and use of film and television work and based on the combination of film its own characteristics. Then the film domain ontology will be built by using the ontology language OWL DLand ontology development environment Protege4.1.In the overall build process, the domain ontology operability, reusability, share ability and scalability are highlighting, which provides an idea to build in other areas of the body.3. Analyzing the reasons for the error that may occur during the process of building film domain ontology. The change the domain knowledge decides the building of film domain ontology is a process of continuous improvement. There will inevitably be all kinds of errors in the process of building, as the ontology of theory and techniques are still in the process of evolving and ontology engineers to understand the depth domain knowledge is not enough. The thesis proposes a method of analysis combined the built film domain ontology by summarizing errors that may appear in the ontology construction process, and classify and organize, analyzing the reasons for each type of error occurs.Through these studies, the thesis will produce the model of Movie ontology, which can solve the current problem of information retrieval systems and recommendation system exists from the semantic level and improve efficiency and quality of service, as well as can greatly satisfy the needs of users; Providing an analysis method based on the Movie Ontology by organizing the errors and their causes, which can provide a reference standard for the ontology engineer.