Article recommendation method and device based on big data statistical analysis

The embodiment of the invention provides an article recommendation method and device based on big data statistical analysis. The method comprises the steps of obtaining multiple pieces of order data; analyzing the order data, and determining a first occurrence frequency of each article in the order data; removing the articles with the first occurrence frequency lower than a first preset threshold value; sorting and combining the remaining articles again, and obtaining a plurality of article sets; determining a second occurrence frequency of each article set in the historical order data; rejecting the article set of which the second occurrence frequency is lower than a second preset threshold value; and determining articles for the to-be-recommended order according to the remaining article set. According to the invention, order data are analyzed, article recommendation is carried out on orders needing to be recommended according to the number of articles purchased by a user in historical orders, and marketing