Design And Implementation Of Intelligent Advertising Material Delivery Management Platform Based On Big Data

With rapid development of the Internet,lots of companies have gradually explored a model of commercial liquidity and gradually applied this model as a core model.This model is known as the marketing discipline of computing advertising.The goal of computing advertising is to maximize the overall benefits of advertising.In computing advertising,Demand-Side Platform(DSP)is a platform for Internet advertisers to provide services for advertisers as well as a platform to implement effective ad auctions.With the continuous application of computing advertising,some typical problems of DSP begun to appear.For example,a large number of fake advertisements which may impair people's thoughts and physical and mental health,the low quality of advertising materials,and the increase of the cost for manual examination of advertising materials.The user experience is getting worse since the proliferation of low—quality advertisements,which could lead to a serious impact on the effectiveness of calculating advertising.There would be a certain risk to the large-scale delivery of advertisements which contain unqualified content such as cheating,fraud,and gambling.These problems have become an urgent problem for Internet companies to achieve their business goal.This thesis focuses on the control of the quality of mass advertising materials,the high costs of manual monitoring of advertising materials,the analysis and implementation of controlling advertising risks.Based on massive data of advertising materials,this thesis proposes an intelligent advertising materials delivery management platform.Through the research on the background of advertisement materials management in the era of big data,this thesis combines the analysis of the status of technology and a unified demand management approach and determines the company's needs and goals.This thesis also designs some modules such as advertising material upload module,a machine audit module,audit client module,creative service analysis module,and administrator management module.Based on the service's architectural model,this thesis achieves intelligent review of advertising materials through machine learning and deep learning.Furthermore through visualization technology,it achieves monitoring and analysis of advertising material quality.More importantly,through streaming computing,it realizes risk monitoring of advertisements delivery.As a result,it will finally achieve the goal of lowering the manual auditing cost,the intelligent control of advertising materials,the control and quantitative analysis of the risk of advertising placement,the quality control of advertising materials,and will improve the efficiency of advertisement material examination.At last,an increase of advertising revenue will be achieved through the process above.