Practice on Personalized Recommendation Strategies Using Deep Learning Algorithms in Online Marketing

With the penetration of big data into various fields of economic life, banks, insurance, transportation, e-commerce, education, healthcare and other fields have generated massive data and information flows, and gradually developed big data sales systems represented by Taobao, JD.com, and others. In this context, “reconstruction” and “subversion” are the key to building personalized marketing systems for retail enterprises. However, personalized marketing is not a new concept. It is based on the modern information way, through the construction of personalized customer service system, so that enterprises can measure and expand channels at a lower cost. This is also the core of Internet marketing. Therefore, in the context of big data, conducting personalized marketing can help industries and enterprises better find target customers, achieve precise management of them, and obtain profits from them. Therefore, this article integrated personalized marketing and customer loyalty in retail enterprises into a common research framework, and explored the impact mechanism of personalized marketing on customer loyalty on this basis. This article constructed a personalized recommendation model by collecting and analyzing user behavior data, combined with deep learning algorithms. The experimental results showed that the click through rate of the experimental group (Group B) in March was <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$r_{B}&gt;55 \%$</tex>, while the highest click through rate of the control group (Group A) did not exceed 60 %. This shows that the recommended model of bipartite projection is progressiveness.