Design And Implement Of Algorithm Library For Variability Modeling Technology

With the gradual increase in the amount of data processing,the gradual improvement in data processing requirements,and the gradual diversification of business development,traditional big data development platforms have been unable to support so many business iteration requirements.The project background of this article is to meet the above requirements.In the big data analysis system,a set of algorithm runtime library that can quickly build personnel analysis algorithm business is developed.The personnel algorithm processing library focuses on solving the problems of high algorithm development threshold,slow business algorithm iteration speed,and complex algorithm association execution in personnel business.In this thesis,combining variability modeling and rule engine technology,a new personnel processing algorithm runtime library is proposed.The algorithm runtime library reduces the dependence of business personnel on algorithm coders through variability modeling technology,and improves business personnel through business algorithm editing and visualization.The development efficiency reduces the coding threshold of business algorithms,and effectively solves the pain points of code developers who frequently modify business algorithms.The main work of this thesis as follows:1.Develop front-end visualization tools.By developing the rule definition function in the rule engine,the visualization of front-end business algorithm editing is realized,and the definition function of the visual syntax tree is provided at the same time,which reduces the use threshold of business personnel and improves the efficiency of business algorithm development.2.Use rule definition and rule engine in the editing and running of business algorithms.The variability modeling technology and rule engine technology are applied to the development of business algorithms.By using the algorithm technology provided by the rule engine,the split business algorithms can be quickly matched to realize the automatic operation of business algorithms.3.Algorithm library provides rule engine execution result call.Through this call,the algorithm library and the business algorithm can be expanded,and the business algorithm logic and code can be separated by calling the operation result of the business algorithm,which reduces the excessive dependence of business developers on algorithm coders.Aiming at the problem of inefficient operation of combination algorithms,this algorithm library can effectively reduce the algorithm configuration time of business algorithms,improve the combination efficiency of business algorithms,achieve the goal of rapid development and deployment of algorithms in personnel processing business scenarios,and improve the combination of personnel and business algorithms.efficiency of execution.In view of the high threshold for business algorithm development,the front-end visualization tool has lowered the threshold for use,and the use of visual syntax tree definitions has lowered the threshold for business personnel to develop business algorithms.For program coders,by providing external calls,the scalability of the system is improved,the workload of coders and the dependence on business personnel are reduced,and the efficiency of business development is effectively improved.