The Application Of Personalized Recommendation Algorithm Based On Micro-blogging In Focus Marketing

Focus Marketing is a breakthrough in traditional marketing, and its emphasis onprecise, subdivision, tangible results. Microblogging is the most popular in recentyears, emerging network media, the speed of its information dissemination, the highdegree of user participation, users of the strong cohesive force, coinciding with theconditions necessary for the Focus Marketing coincide.Focus marketing based microblogging, There are two very important thing is that:firstly,accurate predictions marketing products which users are interested in, and canlead to sustained attention and discussion of these users. Secondly, for those users whoare interested in what manner recommended it be possible to make the microbloggingusers active interest in advertising, trust advertising recommendations, the bestrecommendation.The problem of how to accurately predict the spread of marketing messages, wepropose a model of the concept of the the microblogging system of informationdissemination, of the microblogging information dissemination process described, thethe microblogging information dissemination model mathematical formula, and finallythrough Nutch toolscrawl Sina microblogging data model simulation analysis, andsimulation analysis of the results with the actual topic spread data for comparativeanalysis, come to the topic of this article dissemination of the conclusions of themodel data fit the actual topic propagation model.The way in which the recommended paper analyzed existing recommendationalgorithm based on combined data characteristics of microblogging, collaborativeanalysis and content analysis based on demographics integration a previous fusionrecommended framework recommendedthe algorithm OMCFA, the algorithmconverged user similarity and the similarity of microblogging into the recommendedframework, Bayesian mixed effects regression model, Markov Monte Carlo methods to obtain the values of various parameters in the model utility formula. Algorithmvalidation Finally, the data in the third chapter, and simple collaborative filteringalgorithm, compared to content-based recommendation algorithm obtained OMCFAalgorithm recommended better conclusion.Finally, the paper made a summary of the text, and future research directions aresuggested.