Fast Algorithms for Discovering Sequential Patterns in Massive Datasets

Abstract

Problem statement: Sequential pattern mining is one of the specific data mining tasks, particularly from retail data. The task is to discover all sequential patterns with a user-specified minimum support, where support of a pattern is the number of data-sequences that contain the pattern. Approach: To find a sequence patterns variety of algorithm like AprioriAll and Generalized Sequential Patterns (GSP) were there. We present fast and efficient algorithms called AprioriAllSID and GSPSID for mining sequential patterns that were fundamentally different from known algorithms. Results: The proposed algorithm had been implemented and compared with AprioriAll and Generalized Sequential Patterns (GSP). Its performance was studied on an experimental basis. We combined the AprioriAllSID algorithm with AprioriAll algorithm into a Hybrid algorithm, called AprioriAll Hybrid. Conclusion: Implementation shows that the execution time of the algorithm to find sequential pattern depends on total no of candidates generated at each level and the time taken to scan the database. Our performance study shows that the proposed algorithms have an excellent performance over the best existing algorithms.

Mining Experiential Patterns from Game-Logs of Board Game

“…In this field, many efforts of mining sequential patterns have been devoted to developing efficient algorithms, such as GSP [27], SPADE [28], CloSpan [29], PrefixSpan [30], and MEMISP [31]. In recent years, researchers have paid more attention to the applied research of sequential pattern mining as discussed elsewhere [32][33][34][35][36].…”

“…(16) for each allFC in vAllFC do (17) for each 𝑐 in allFC do (18) if allFC.index == (MDab − 1) then (19) isBreak = true; break; (20) endif (21) if 𝑐 ⋅ 𝑥 == 𝑃 𝑏 ⋅ 𝑥 and 𝑐 ⋅ 𝑦 == 𝑃 𝑏 ⋅ 𝑦 then (22) Dab = allFC.index + 1; isBreak = true; (23) break; (24) endif (25) if allFC.index == (MDab − 2) then break; endif (26) FindVirtualFallingCheckers(0, CS, 𝑐 ⋅ 𝑥, 𝑐 ⋅ 𝑦, true, vAllFC, allFC.index + 1); ( 27) endfor (28) if isBreak == true then break; endif (29) endfor (30) return Dab; (31) //Search the virtual falling checkers of specified checkers. (32) Function FindVirtualFallingCheckers(𝑑, CS, 𝑥, 𝑦, isFirstStep, vAllFC, index) (33) begin (34) for 𝑖 = 3 to −3 do (35) if 𝑖 == −𝑑 then continue; endif (36) if 𝑖 == 0 then continue; endif (37) Search the next jumping falling checker (𝑥…”