Intelligent social advertisement release platform based on portrait data of big data

Abstract

The invention provides an intelligent social advertisement release platform based on portrait data of big data. Real-time and offline behavior data of a user can be collected by a single interface (based on a dataflow model); integration and calculation are performed through a big data component; a calculation result is user portrait data; and a data organization form is a wide-table taking tags as fields. The portrait data is organized in a reverse index mode; and indexes of the data are stored in a big data warehouse. A Web platform takes multiple specific tags as input conditions and outputs statistical and aggregation results. During use, data specifically used for interests, hobbies and the like is analyzed, and advertisement release is performed according to specific hobbies of the user, so that the intelligent social advertisement release platform is formed, the advertisement release randomness and blindness are reduced, and the pertinence is achieved.

Description

technical field [0001] The present invention relates to the technical field of an intelligent social advertisement delivery platform, in particular to an intelligent social advertisement delivery platform based on portrait data of big data. Background technique [0002] At present, on the one hand, advanced advertisements are usually placed in the following ways: one is through physical advertising screens, such as outdoor billboards and advertising display boards, and the other is through specific web pages on the Internet, inserted between TV programs For example, all major media websites will embed advertisements, and various satellite TV programs and online videos will also insert advertisements for a certain period of time. These are all forms of placing advertisements among viewers. Therefore, even if it is disgusted by the audience, the effect of advertising on product marketing is not very good, and even if popular stars are invited, it cannot effectively improve the...

AI Technical Summary

Problems solved by technology

[0008] The deficiencies in the existing technology include: first, the existing user portrait technology rarely involves computing operations with a large amount of data, and most of them are based on traditional relational databases. If the amount of data is large, the calculation will fail or the calculation will fail. too long

Third, even if the real-time calculation method is applied, the data collection and calculation are heterogeneous and complex, and there is no unified engineering method to simplify and automate the whole process of offline and real-time calculation

Fourth, the existing user portrait query technology is mainly the detailed query of portrait data, and the storage query technology is also limited to relational databases, and does not retrieve and analyze user portrait data from the perspective of multi-dimensional analysis

Fifth, the existing multi-dimensional analysis systems are mainly ROLAP and MOLAP. The architecture between the systems is complex and difficult to maintain. It is necessary to establish a data cube structure or a new view, which also leads to performance loss in retrieval.

Images