Personalized recommendation method and system for reading content based on AI open platform

Abstract

The invention discloses a personalized recommendation method for reading content based on an AI open platform, including: receiving personal face images taken by users, sending the personal face images to the AI ​​open platform for face recognition, and receiving feedback from the AI ​​open platform Face recognition results, match the face recognition results with the pre-stored user registration face image recognition results in the user database, if the match is successful, it means that the user has successfully logged in; personalized recommendation: receive the image of the current reading article taken by the user, Send the image of the currently read article to the AI ​​open platform for text recognition, receive the text recognition result fed back by the AI ​​open platform, and then send the text recognition result to the AI ​​open platform to extract the article label, receive the article label fed back by the AI ​​open platform, and upload the article The tags are matched with the book information database, and the matched book names and the book names with the highest word meaning similarity to the article tags are output.

Description

technical field [0001] The invention relates to a reading content personalized recommendation method and system based on an AI open platform. Background technique [0002] With the advancement of technology and the popularization of electronic products, people rely more and more on convenient and fast smart products in daily life. In recent years, smartphone brands have continued to expand and their functions have gradually improved. According to the statistics of Netease Technology, 98% of post-90s college students use smartphones, and apps for socializing, shopping, games, and reading news are the main force in mobile phones of young users. The top three must-have apps for mobile phones of post-90s students are: social networking, leisure and entertainment, and study and work. Among them, learning and working software has great development potential in the market, so opening up such software is in line with the mainstream and has great practicability. [0003] College s...

AI Technical Summary

Problems solved by technology

[0004] First, the existing reading content recommendation method relies on the software that comes with the mobile terminal itself. The software takes up a lot of memory, and the recommendation calculation process is slow, and cannot quickly respond to the user's recommendation request;

[0005] Second, the existing reading content recommendation methods are often for commercial sales needs, and cannot achieve personalized recommendations. The recommendation accuracy is not high, the academic relevance is not high, and the recommended results are often not books that college students care about.

Images