Design of Personalized Recommendation Model for Precision Marketing Based on Collaborative Filtering Algorithm

In recent years, the application of e-commerce has become more and more extensive, and the current society has gradually entered the era of big data. Based on this background, personalized recommendation of e-commerce precision marketing has also become the focus of attention in society. The concept of “new retail” has enabled the retail industry to find a new development direction and driving force, and the concept of “data-driven” has become a guide for the deep integration of online channels and offline channels. Commercial banks, which have a similar commercial structure to the retail industry, are also facing the common problem of “retail” in which offline channel resources are developed but the cost is high, and the development of online channels is gradually accelerated but the experience is insufficient. In order to achieve precise delivery of specific specifications of products and specific retailers, this article proposes a retail oriented sales volume prediction model based on collaborative filtering(CF) recommendation system technology - SPM-CF model, which is innovative in applying recommendation system technology to the traditional retail field rather than the e-commerce field, The algorithm of sales volume score transformation is studied to solve the problem of how to compare and analyze retailers of different sizes. Using local retail order data for offline testing, the effectiveness of the model was verified, and the applicability of the model to specifications that have been fully promoted in some regions was explained.