Open AccessResearch on user behavior recommendation algorithm for business process
Author(s) -
Kunying Li,
Dingwen Yu,
Xiaolian Li,
Ying Song,
Wenting Zhang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1570/1/012013
Subject(s) - computer science , similarity (geometry) , process (computing) , apriori algorithm , relevance (law) , sorting , data mining , node (physics) , set (abstract data type) , algorithm , matching (statistics) , recommender system , a priori and a posteriori , business process , business process discovery , table (database) , information retrieval , association rule learning , business process modeling , artificial intelligence , work in process , mathematics , engineering , philosophy , operations management , law , image (mathematics) , structural engineering , operating system , epistemology , political science , programming language , statistics
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.