Method and system for analyzing association between network advertisement audience behaviors and audience interests

Abstract

The invention discloses a network advertisement audience behavior and audience interest association analysis method and system. The method comprises the following steps: firstly, reading an audience identity model from a crowd database or (new audiences without historical data) to generate a new identity model through a similarity forest algorithm; and then updating (or assigning values for the first time) the audience model according to the browsing track, frequency, content and other data of the current behavior of the audience, performing feature weight comparison with the population feature model and the interest model, and pre-judging the population feature and interest based on the score of each attribute weight. And meanwhile, the data generated by the method is fed back to the related functional model, and iterative updating is continuously carried out, so that the whole functional system is more stable and effective than other existing products.

Description

technical field [0001] The present invention relates to the field of Internet application technology, in particular to an analysis method and system for analyzing the relationship between audience behavior and audience interest in Internet advertising in a series of Internet-related industries such as Internet digital marketing industry, big data mining and analysis industry, and Internet of Things industry . Background technique [0002] According to the 41st "Statistical Report on Internet Development in China" released by the China Internet Network Information Center (CNNIC) on January 31 this year, as of December 2017, the number of Internet users in my country has reached 772 million, with new additions throughout the year. There are 40.74 million netizens, and the scale of netizens has maintained a steady growth. Among them, the number of mobile Internet users reached 753 million, accounting for 97.5% of the total Internet users. At the same time, the number of Interne...

AI Technical Summary

Problems solved by technology

The interest classification, age, and gender filled in the visitor's personal account have better adaptability to long-term individual audience analysis, but lack of tracking of long-term interest changes of the audience, and are determined by the state of the audience when they filled in before, not Verified by the objective and actual content, the authenticity of the audience’s filling cannot be ruled out

[0006] 2. Classify the interests of the website content that the visitor is currently visiting. This method focuses on the current instantaneous interest of the visitor. The attribute and interest classification of the visitor is derived from the content of the visit. It is more advanced than the first method, but the disadvantage is that it cannot Completely reflect the long-term interests of visitors. It is impossible to analyze the interest ratio of short-term visits in the long-term state, and it is impossible to distinguish the true intentions of visitors with time curves.

At the same time, affected by the content analysis of the page, the filtering of negative information and the transmission of positive information cannot be well identified

[0007] 3. The third method is to combine long-term visit interest with the current page, which is more advanced than the first two, but does not consider the impact of time curve and frequency on interest

Images