Research on user behavior recommendation algorithm for business process

Abstract

In this paper, ProcessLink is a user behavior judgment and recommendation algorithm based on the information system business process. The purpose is to provide users with reliable unknown process selection. The algorithm idea is as follows: firstly, collect and analyze historical data, establish the relevance feedback set of “Time-Role -Process”, establish the interest relationship decision table of each role according to the process flow link behavior between users. Starting from each process node, the Apriori algorithm completes a round of initial filtering on these relationships, and obtains the similarity association degree of each role node in the process link; finally, the recommendation sorting is carried out through the recommendation algorithm model matching and scenario similarity calculation. This research realizes the accurate prediction of the user’s next link and gives the corresponding opinions. The experimental results show that the algorithm is effective and accurate in the business process.