The Research On Personalized Recommendation Of Information Distribution Platform

The development of big data in the Internet age has made algorithmic technology nested in all areas of life.In the context of artificial intelligence,news media industry has also undergone many changes in the business chain and media formats.The rise of the personalized recommendation-based information distribution platform solves the problem of accurate adaptation of information and people,innovates people's information behavior model,realizes content-based user aggregation,avoids information selection anxiety in the Internet age,and recommends the algorithm itself.Limitations also bring about many drawbacks.This article takes the "distribution of information" of the information distribution platform as an example to clarify the causes of personalization of the information distribution platform by reading the leading-edge literature on "individualized recommendation" in Chinese and Western media research and combining its own practical observations on algorithm recommendation services.The recommendation mechanism sorts out the multi-dimensional value innovation of personalized recommendations for news formats and the difficulties it faces,and puts forward optimization proposals in terms of technology,checks,and content.The study was conducted from six aspects.The first part is the introduction,mainly from the background of personalized recommendation,literature review,research innovation,research methods and other aspects.In the second part,starting from the background of information anxiety,the motivation of the rise of personalized recommendation is summarized as follows:the upgrading of user information needs,the pursuit of long-tail effect of commercial capital,and the support of algorithmic technology.The third part discusses the new representation of the new media evolution process and the personalized trend embodied in the personalized recommendation under the framework of the media compensation theory.The fourth part revolves around the"tagging" to unlock the personalized recommendation mechanism from the platform,text,and user perspectives.The fifth part summarizes a number of innovations that are personalized and recommended based on the standpoint of the news business chain and news ecological innovation.The sixth part analyzes the development dilemma of personalized recommendation from the limitations of personalized recommendation algorithm,the information problems brought by personalized information,the problem of democratic governance,and the hazard of content value.The last part puts forward suggestions based on the status quo of development,targeted from the aspects of computational optimization,human-machine collaboration,and rich content production subjects.This study believes that the personalized recommendation reconstructs the research of traditional information production and distribution and brings about the transformation of the traditional information dissemination paradigm.Personalized recommendations have overturned our knowledge of news texts,audience information acceptance behaviors,and news production,and have brought about an expansion of the news field.Western research has recommended algorithms as a kind of social governance and also demonstrated its significance to social construction.However,the shift in the algorithm recommendation between China and the West still highlights the limitations of the algorithm and the series of risks posed by the lack of the gatekeeper.Improvements must be made in terms of technology,human resources and content.