The invention discloses a
web service recommendation method based on theme and service combination information. The method includes the steps of firstly, obtaining a historical
score record of user web services and description information and service combination information of the web services; secondly, obtaining a service-theme matrix through an LDA model, and calculating the theme similarity between the web services according to the service-theme matrix; thirdly, calculating a service combination
similarity matrix between the web services; fourthly, selecting a neighbor set of each
web service according to the theme similarity and the service combination
similarity matrix; fifthly, fusing the neighbor set and the
similarity matrix with an SVD
matrix decomposition model to predict a store; sixthly, selecting a plurality of services with the highest
score according to the predicted
score for recommendation. In the method, by introducing the considering of the service theme informationand the service combination information, the new dimension is introduced to
web service recommendation, the recommendation precision is improved, and meanwhile the method can solve the
cold start problem of partial service recommendation and has wide market prospects.