IP positioning method based on Bayes and deep neural network

Abstract

The invention discloses an IP positioning method based on Bayes and a deep neural network. The IP positioning method based on the Bayes and the deep neural network comprises an IP equipment comprehensive positioning method based on social information, an IP positioning method based on weighted naive Bayes and an IP positioning method based on a BP neural network. Compared with the prior art, the unknown IP is comprehensively positioned by fusing three methods, and corresponding methods can be adaptively selected for positioning according to different collected data types.

Description

technical field [0001] The present invention relates to the field of network technology, in particular to an IP positioning method based on Bayesian and deep neural network. Background technique [0002] The goal of IP-based network entity geolocation technology is to locate the geographical location of users and devices in the network under the condition of knowing the IP addresses of users or devices. Therefore, IP geolocation technology can help Internet service providers distinguish users geographically, so as to provide more geographically related services. [0003] For example, for targeted advertisements, when host users or mobile device users visit web pages and APP applications, if the Internet service provider knows the user's location, it can add more targeted advertisements based on the user's geographical location, thereby improving How well your ad served. Some websites and applications can also adjust the page content according to the user's geographical loc...

AI Technical Summary

Problems solved by technology

[0005] The first category of registration information query includes using a Whois-like database to query host IP location information or query domain name information. The disadvantage of this type of method is that the database records the information of registered users, which may be different from the real location of IP. There are deviations, and the data update is slow, resulting in low data accuracy

[0006] The second type of web page information extraction, such as mining the IP used by the user, mining the address information that the user filled in when registering, or the address information selected when using the APP application, or through the phone, fax, address, etc. contained on the web page Information to collect website IP and corresponding geographic location. The limitation of this method is that the obtained IP coverage is small, and at the same time, it is impossible to judge the validity of the information submitted by the user, and it is impossible to avoid false information, and it may violate the privacy of the user.

[0007] The third type of network measurement. For example, the GeoPing algorithm believes that the round-trip delay between hosts in similar networks is strongly similar. The disadvantage of this algorithm is that the positioning result is limited by the position of the reference point, and the error is greater in places with fewer reference points. , such as the constraint-based CBG algorithm and TBG algorithm, their limitation lies in the relationship between delay and physical distance, but the delay is greatly affected by the network topology and network environment, so the relationship between delay and physical distance is not clear

Images