Research On Identification Method Of Civil Aviation Master Data Based On Internet Data

As the scope of civil aviation business is becoming wider and wider,there are a series of common data between different systems,services,and even different departments.Because the data naming and storage rules are inconsistent between various business departments and subsystems,the information between the systems is asymmetric,resulting in a large amount of data redundancy,and the airline's rich operational data cannot be fully utilized.At the same time,passengers' demand for information is not limited to relevant aviation data,but more hopes to obtain more comprehensive non-aviation Internet data such as hotels and tourism.Therefore,in order to improve the quality of civil aviation master data and help airlines improve operational efficiency,There is an urgent need to study civil aviation master data identification methods that integrate Internet data.Firstly,in order to integrate non-aviation Internet data such as hotels and tourism,and to solve the problem of data inconsistency when the same entity is referenced by multiple data sources,an entity matching method based on non-primary attribute outlier detection is proposed.The idea is to extract public non-primary attributes from different data sources,and sort according to their importance to select appropriate non-primary attribute sets.Combine the outlier detection model to filter the entity pairs,and then use machine learning methods to select the appropriate ones.The matcher trains to get the matching pair.Secondly,in order to overcome the shortcomings of the single weighting method brought by the traditional master data identification method,a civil aviation master data identification method based on combined weight-cloud model is proposed.According to the characteristics of the main data,the scoring index of the civil aviation master data is considered and its rating criteria are established.The information entropy is introduced on the basis of the improved analytic hierarchy process,and the comprehensive weight is calculated.Then the cloud model is used to calculate the membership degree of the sample.Finally,through experiments on non-aviation and civil aviation data,the results show that the proposed method can effectively integrate non-aviation Internet data and identify the main data,which overcomes the shortcomings of traditional entity matching that cannot be applied to large-scale data.Identification provides a new solution.