Ontology-based Offshore Case Recommendation System Research And Implementation

With the development of industrial information technology,this technology has been widely used in various work scenarios of offshore engineering equipment enterprises,and there have been many information systems with offshore applications as the main application scenarios.However,these case information systems generally have problems of low information utilization and weak individualized information services,which greatly affects the experience of users within the enterprise.Therefore,the thesis studies the ontology-based text similarity calculation method and the recommended method of offshore engineering case,and proposes an ontology-based offshore case recommendation method.The main research work is as follows:(1)Since there is no ontology in the field of marine engineering equipment,the article combines the resources of the existing offshore equipment field and the ontology construction process of the seven-step method,in order to excavate the semantic information contained in the knowledge of offshore equipment,and constructs the offshore equipment field with Protégé software.Ontology.(2)Based on the existing ontology-based semantic similarity calculation method,this paper combines other factors that affect semantic similarity,and proposes an improved conceptual semantic similarity calculation method.By introducing the extended query technology,the paper solves the problem of the missing semantic information of the existing text similarity calculation,and proves that the improved text semantic similarity calculation method has higher accuracy than the traditional text semantic similarity calculation method.(3)The article analyzes the usage of users in reality,and designs a recommendation algorithm based on the concept of offshore engineering and a recommendation algorithm based on the concept of user interest.In order to evaluate the performance of these two recommended algorithms,the paper designed a comparison experiment to compare these two algorithms with the classic TF-IDF algorithm.The experimental results show that the accuracy of the recommendation method based on the similarity of the offshore case is better than the recommendation method based on the similarity between users,but the recall rate is lower than therecommendation method based on user similarity.Moreover,both algorithms have better performance than the TF-IDF algorithm.(4)According to the proposed similarity calculation method and the marine case recommendation method,the ontology-based offshore case recommendation system is designed and implemented using the offshore equipment domain ontology.The system can tap the potential interest of the user while satisfying the basic needs of the enterprise and recommend the offshore case for the user according to different situations to realize personalized recommendation.