Short message review system and method and method for constructing short message review model

Abstract

A short message review system and method and a method for constructing a short message review model, the review system includes a URL detection module, an error correction matching module, a URL black and white list module, a URL content module, a real-time capture module, a review model, a model training module and a background management module. The audit method is to use crawler technology to capture the URL in the text message in real time, then combine the text message content and web page content to model at the same time, and use the machine learning algorithm to classify the nature of the text message content to be audited, and finally determine whether the text message needs to be released or blocked. . The accuracy of the present invention is greatly improved compared with the traditional audit method based on the content of the short message and the URL itself, especially when auditing the URL shortened by the short URL service, the traditional method can only be one size fits all, but the present invention can be based on the URL after the jump Different URLs for personalized review.

Description

technical field [0001] The present invention relates to the field of information review, in particular to a short message review system and method and a method for building a short message review model, that is, a review system and method for including URL links and other jumpable information in a piece of information and a method for building a short message review model . Background technique [0002] In recent years, the number of users of smart mobile phones in my country has grown at an unimaginable rate, and due to some essential characteristics of smart phones, people's usage habits have changed dramatically. For example, when using non-smart phones before, if business users need to Sending a website address to a personal mobile phone and allowing the user to click to open it can only use the WAPPUSH technology. If the SMS technology is used, the user who receives the SMS can only copy the link address to the browser and open it. [0003] Due to WAP PUSH's own usage h...

AI Technical Summary

Problems solved by technology

[0006] 2. Send the activity address in the member marketing SMS, such as: Dear XXXX users, the 2017 GL82.0T Automatic Deluxe Edition is not yet on the market, and the official guide price has not yet been announced

[0009] The above are just some typical applications. In fact, there are more and more SMS applications with URLs, and the content is also varied. With the popularity of this application, it also brings a severe test to the current information review management: due to the existence of URLs Certain concealment, the webpage address pointed behind the specific URL cannot be known in the content of the SMS, and even due to the character limit of the SMS, most of the URLs appearing in the SMS are short addresses, which are shortened by professional short URL services Yes, and even if the content is the same, the URL sent in each text message can be different, such as t.cn in the example, resulting in the inability to check the URL address library

These problems have brought difficulties to information review, resulting in a greatly increased risk of information review

[0010] In the traditional information review method, the URL or part of the information in the URL is often put into the address library, and when the text message content includes the information in the URL address library, it will be released or blocked according to the results recorded in the library, but This audit method is almost impossible to work properly due to the emergence of short URL services, such as the short URLs such as http: / / t.cn / ZtG0V2G and http: / / 2bai.co / 969644 in the above example, The domain name itself does not have characteristic elements, and the latter part is randomly generated, so in the traditional information review system, these short messages are either all intercepted or all are released, and finally the URL review module is useless

[0011] Another most stupid method is that the SMS gateway with a manual review module will send the SMS containing URL keywords to the letter review personnel for manual review. The way to achieve it requires a lot of manpower and material resources, and at the same time greatly increases the delay of SMS messages and increases the company's operating costs

Images