A Kind of Personalized Advertising Recommendation Method Based on User-Interest-Behavior Model

With the rapid development of internet and mobile internet, delivering ads on the internet has become the main channel for major advertisers. However, the accurate recommendation of online ads is a major problem which plagued advertisers and agencies. By analyzing the unstructured characteristics of online ads and search engine user behavior data, proposed a kind of personalized ads recommendation method based on User-Interest-Behavior model, which can extract the user’s interest preferences by the topic model and generate the recommended list of ads based on the nearest neighbor and user behavior. The experimental results demonstrate that the personalized ads recommendation method based on nearest neighbor and user behavior can recommend personalized ads and have a better performance than the content-based recommendation method.