With the development of information technology, a large amount of data is generated and correlated. Realizing the integration of data from different channels, comprehensive analysis and data mining of multi-source data have gradually become a significant research issue. As the needs of various types of talent recruitment increases, website recruitment has been universal, along with multi-source data environment. It is a typical scenario for data fusion and data mining. The research proposes data comprehensive mining framework based on data fusion and machine learning methods. After that, taking website data analysis as an example, the research integrates various open-source data such as website data, city developing data, time data and other relevant data as well as displays the distribution characteristics of website recruitment of talents. It systematically excavates the inherent law of website recruitment of talents, identifies the factors affecting position popularity, and puts forward policy suggestions. It provides the theoretical reference and case application for the comprehensive framework of data fusion and data mining.