Abstract
Extracting useful information from huge amount of data is known as Data Mining. It happens at the intersection of artificial intelligence and statistics. It is also defined as the use of computer algorithms to discover hidden patterns and interesting relationships between items in large datasets. Candidate generation and test, Pattern Growth etc. are the common approaches to find frequent patterns from the database. Incremental mining is a crucial requirement for the industries nowadays. Many tree based approaches have tried to extend the frequent pattern mining as an incremental approach, but most of the research was limited to interactive mining only. Here, instead of tree based approach, graph based approach is presented which also gives good results for incremental mining. Â
Highlights
Industries suffer the problem of huge amount of data
Nowadays incremental mining is in high demand because of the continuously increasing size of the database
After day by day updation of the database if the mining procedure starts from the scratch, it is useless
Summary
Industries suffer the problem of huge amount of data. It is very difficult to find out which information is important and useful in day to day life. FP-growth algorithm has an extended prefix-tree structure for storing compressed, crucial information about frequent patterns for mining the complete set of frequent patterns by pattern fragment growth This graph based concept works basically on the FP-Tree and FP-Growth approach In this approach the first scan of the database finds the frequent items. On the base of the first scan of the database, graph will be formed which includes only frequent items It creates the same problem as FP-growth, means the mining procedure has to be started from the scratch. After that Vivek Tiwari et al (2010) proposed another graph based approach for association rule mining It works mainly on three phases: The first is to generate the graph including all items.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
