There are many past studies will be applied to data mining patents on theclassification of documents, while ignoring the patent information itself usefulinformation, if there will be patent application of these important information to thepatent documents than the recommended, may provide a better The patent documentsthan the method. Therefore, the study of social network analysis, and use ofbibliographic information in patent documents of identity, to try to improve therecommendation of the accuracy of the patent documents. To prove that the mainpurpose of this study was to give future researchers and to create a separate analysisof the patent subject. Therefore, the main conclusions of this study are: (1) to socialnetwork-based system of patent documents can recommend the recommendation ofpatent documents to the user. (2) to patent law books for the community networknodes in the patent documents link between the intensity settings, the existence of the best books of the patent portfolio.