Report

Mining Frequent Patterns Using FP-Growth Method Ivan Tanasić ([email protected]) Department of Computer Engineering and Computer Science, School of Electrical Engineering, University of Belgrade Mining Frequent Patterns without Candidate Generation: A Frequent-Pattern Tree Approach ◦ ◦ ◦ ◦ Jiawei Han (UIUC) Jian Pei (Buffalo) Yiwen Yin (SFU) Runying Mao (Microsoft) Ivan Tanasic ([email protected]) 2/25 Problem Definition Mining frequent patterns from a DB ◦ Frequent intemsets (milk + bread) ◦ Frequent sequential patterns (computer -> printer -> paper) ◦ Frequent structural patterns (subgraphs, subtrees) Ivan Tanasic ([email protected]) 3/25 Problem Importance 1/2 Basic DM primitive Used for mining data relationships ◦ Associations ◦ Correlations Helps with basic DM tasks ◦ Classification ◦ Clustering Ivan Tanasic ([email protected]) 4/25 Problem importance 2/2 Association rules ◦ buys(“laptop”)=>buys(“mouse”) [support = 2%, confidence = 30%] • Support=% of all transactions containing that items •Confidence=% of transactions containing I1 that contain I2 Ivan Tanasic ([email protected]) 5/25 Problem Trend Apriori speedup using techniques New data structures (trees) Association rule specific algorithms Specific AR algorithms (OneR, ZeroR) FP-Growth still widely used Ivan Tanasic ([email protected]) 6/25 Existing Solutions 1/3 (Apriori) Agrawal et al. (1994) AP: All nonempty subsets of a frequent itemset must also be frequent Starts from 1-itemsets Join + prune (using AP + min supp) Generates huge number of candidates Ivan Tanasic (i[email protected]) 7/25 Existing Solutions 2/3 (ECLAT) Zaki (2000) Equivalence CLass Transformation Vertical format: {item,TID_set} instead of {TID,itemset} Intersects TID_sets of candidates TID_sets holds support info (no scans) Still generates candidates Ivan Tanasic ([email protected]) 8/25 Existing Solutions 3/3 (TreeProjection) Agarwal et al. (2001) Creates a lexicographical tree and projects db into sub-dbs based on the patterns mined so far Recursively mines subdatabases Less scalable then FP-Growth Ivan Tanasic ([email protected]) 9/25 FP-Tree construction 1/6 Desc. supp. sort • Min support = 2 Ivan Tanasic ([email protected]) 10/25 FP-Tree construction 2/6 Desc. supp. sort T1={I2,I1,I5} Ivan Tanasic ([email protected]) 11/25 FP-Tree construction 3/6 Desc. supp. sort T1 = {I2, I1, I5} T2 = {I2, I4} Ivan Tanasic ([email protected]) 12/25 FP-Tree construction 4/6 Desc. supp. sort T1 = {I2, I1, I5} T2 = {I2, I4} T3 = {I2, I3} Ivan Tanasic ([email protected]) 13/25 FP-Tree construction 5/6 Desc. supp. sort T1 = {I2, I1, I5} T2 = {I2, I4} T3 = {I2, I3} T4 = {I2, I1, I4} Ivan Tanasic ([email protected]) 14/25 FP-Tree construction 6/6 Desc. supp. sort T1 = {I2, I1, I5} T2 = {I2, I4} T3 = {I2, I3} T4 = {I2, I1, I4} T5 = {I1, I3} T6 = {I2, I3} T7 = {I1, I3} T8 = {I2, I1, I3, I5} T9 = {I2, I1, I3} Ivan Tanasic ([email protected]) 15/25 Mining of the FP-Tree 1/4 It. Conditional P. base Cond. FP-Tree Freq. Patterns Generated I5 {I2:2, I1:2} {I2,I5:2},{I1,I5:2},{I2,I1,I5:2} {{I2,I1:1},{I2,I1,I3:1}} Ivan Tanasic ([email protected]) 16/25 Mining of the FP-Tree 2/4 It. Conditional P. base Cond. FP-Tree Freq. Patterns Generated I5 {{I2,I1:1},{I2,I1,I3:1}} {I2:2, I1:2} {I2,I5:2},{I1,I5:2},{I2,I1,I5:2} I4 {{I2,I1:1},{I2:1}} {I2:2} {I2,I4:2} Ivan Tanasic ([email protected]) 17/25 Mining of the FP-Tree 3/4 It. Conditional P. base Cond. FP-Tree Freq. Patterns Generated I5 {{I2,I1:1},{I2,I1,I3:1}} {I2:2, I1:2} {I2,I5:2},{I1,I5:2},{I2,I1,I5:2} I4 {{I2,I1:1},{I2:1}} {I2:2} {I2,I4:2} I3 {{I2,I1:2},{I2:2},{I1:2}} {I2:4,I1:2},{I1:2} {I2,I3:4},{I1,I3:4},{I2,I1,I3:2} Ivan Tanasic ([email protected]) 18/25 Mining of the FP-Tree 4/4 It. Conditional P. base Cond. FP-Tree Freq. Patterns Generated I5 {{I2,I1:1},{I2,I1,I3:1}} {I2:2, I1:2} {I2,I5:2},{I1,I5:2},{I2,I1,I5:2} I4 {{I2,I1:1},{I2:1}} {I2:2} {I2,I4:2} I3 {{I2,I1:2},{I2:2},{I1:2} {I2:4,I1:2},{I1:2} {I2,I3:4},{I1,I3:4},{I2,I1,I3:2} I1 {{I2:4}} {I2:4} {I2,I1:4} Ivan Tanasic ([email protected]) 19/25 How much batter is it 1/3? Runtime on sparse data: Ivan Tanasic ([email protected]) 20/25 How much batter is it 2/3? Runtime on mixed data: Ivan Tanasic ([email protected]) 21/25 How much batter is it 3/3? Compactness: Ivan Tanasic ([email protected]) 22/25 Is it Original? A lot of methods try to improve Apriori ◦ ◦ ◦ ◦ Hashing Transaction reduction Partitioning Sampling TreeProjection uses similar structure, but it is still a different method Ivan Tanasic ([email protected]) 23/25 Importance over time Basic primitive (strong foundation for tall building) Performance gets very important as databases are getting huge Scalability also FP-Growth has both performance and scalability Ivan Tanasic ([email protected]) 24/25 Conclusion An important method for solving important DM tasks Fast Compact Scalable (db projection/tree on disk) Ivan Tanasic ([email protected]) 25/25 Mining Frequent Patterns Using FPGrowth Method Ivan Tanasić ([email protected]) Department of Computer Engineering and Computer Science, School of Electrical Engineering, University of Belgrade