The invention discloses a potential customer recommendation
system and recommendation method based on enterprise tags. The
system comprises a
customer information management module, a customer tag management module, a customer recommendation module and a customer management module. The recommendation method comprises the steps of data
crawling and storage, data cleaning,
label extraction and cleaning, user labeling, graph
data structure construction, calculation of relevancy of a
client vertex relative to a user vertex, calculation of recommendation accuracy, selection of a trained data modelwith high accuracy to be deployed on a
server, and
client recommendation. According to the
system and the method disclosed by the invention, the enterprise
label is extracted more comprehensively by utilizing a distributed
big data processing technology, so that the enterprise is described more comprehensively by using the
label, the relationship between the interest label of the user and the enterprise label is established more accurately, and a recommendation result is more accurate.