Method and apparatus for obtaining and analyzing data information aimed at data object

Abstract

The invention provides one data subject based data information get and analysis method and device, which comprises the following steps: receiving first user trigger to get data subject information composed of user IP address; second first and second user telephone number to inform network connection to first and second user to fulfill dialogue; getting the telephone behavior data information composed of number and valid time length; analyzing IP address relative number and when certain IP address user valid dialogue length is less than preset time, then this trigger behavior is invalid.

Description

technical field [0001] The invention relates to the field of data processing, in particular to methods and devices for acquiring and analyzing data information of data objects. Background technique [0002] At present, with the application of Internet technology more and more widely, many people's daily work and entertainment are carried out on the Internet, and the Internet also provides various services in order to meet the growing needs of the people. The Internet exists in the form of data objects, such as news, multimedia content, online advertisements, etc. In order to enjoy these services, people need to perform various operations on the provided data objects. These operations are called targeting data objects. This not only facilitates the vast number of Internet users, but also brings unlimited business opportunities for various service providers. However, with the development of Internet technology, there are more and more factors that hinder its progress, such as ...

AI Technical Summary

Problems solved by technology

[0004] Although the anti-fraud click technology like Yahoo can identify click fraud to a certain extent, because it is only identified on the basis of network behavior analysis of clicking on online advertisements, such technology has the following defects: First of all, because this type of technology analyzes the user's network behavior completely, it cannot truly and completely identify click fraud, because fraudsters can simulate ad clicks on the Internet completely according to the operation behavior of real users, which is impossible for this type of technology. identified

Secondly, the calculation method is complex, requiring the accumulation of a large amount of user data for analysis, and the cost is high. For example, the core technology AI pattern recognition and filtering function in Yahoo’s anti-fraud system requires the establishment of a huge sample library of fraudulent clicks / invalid clicks around the world. , to train the sample library from all over the world, so that the AI ​​pattern recognition system can accurately identify fraudulent clicks / invalid clicks by integrating multiple features.

Finally, the false recognition rate is high, and some real advertising behaviors are often identified as fraudulent behaviors, which will cause the advertising provider to bear a large amount of losses

Images