SPMF - Contributors

Contributors

Project leaders

  • Prof. Philippe Fournier-Viger (founder) - Shenzhen University, Shenzhen, China
  • Prof. Jerry Chun-Wei Lin - Western Norway University of Applied Sciences, Bergen, Norway
  • Prof. Wei Song - North China University of Technology, Beijing, China
  • Prof. Vincent S. Tseng - National Chiao Tung University, Taiwan
  • Prof. Ji Zhang - University of Southern Queensland, Australia

Algorithms

  • Philippe Fournier-Viger
  • Jerry Chun-Wei Lin, Ting Li, Lu Yang, Shi-Feng Ren, Jiexiong Zhang, et al: implementation of the following algorithms: HAUI-Miner, EHAUPM, HAUI-MMAU, FFI-Miner, MMFI-Miner, SFUPMiner, HUIM-BPSO, HUIM-BPSO-tree, HUIM_GA, HUIM-GA_Tree, IHAUPM, MEMU, SPHUIM, UHUSPM, and PHUSPM and others
  • Wei Song, Chaomin Huang, Junya Li, Lu Liu, Wei Ye, Jiakai Nan, Chuanlong Zheng- implementations of the HUIF-GA, HUIF-PSO, HUIM-ABC, HUIM-SPSO, HUIM-AF, SFUI-UF, SFU-CE, HAUIM_GMU, TKU-CE, TKU-CE+, HUIF-BA, HUIM-ACO, SPM-FC_L, SPM-FC_P, MRI-CE and other algorithms
  • Youxi Wu et al. - implementations of several sequence pattern mining algorithms such as NOSEP, MAPD, NWP-Miner, OWSP-Miner, and others to be released soon
  • Antonio Gomariz Penalver - implementations of the following algorithms: Clasp, CloSpan, Spade (regular and parallelized version), GSP, SPAM (alternative implementation) and PrefixSpan (alternative implementation), and other algorithms..
  • Peng Yang - implementation of SPP-Growth, LPP-Growth, EMMA, MINEPI, MINEPI_PLUS, LPPM(Breadth), LPPM(depth), MRCPPS, HUE-SPAN and others
  • Vikram Goyal, Sonam Rathore, Ashish Sureka, Dhaval Patel, Siddharth Dawar - implementation of several algorithms: UFH, SkyMine, TUP, FHMDS, FHMDSNaive, HMiner, UP-Hist
  • Yangming Chen - implementations of the POERM, POERMH, POERM-ALL algorithms for partially-ordered episode rule mining and algorithms to generate episodes rules with EMMA and MINEPI+, and also the TKP-ENUM and TKP-SCAN algorithms.
  • Ted Gueniche - implementations of sequence prediction models : CPT+, CPT, PPM, DG, AKOM, TDAG, LZ78
  • Cheng-Wei Wu, Jiantao Wang, Yipei Lai et al. . - implementations of several algorithms and some code also published in UP-Miner, under the GPL license including CHUD, TKU, and US-SPAN.
  • M. Saqib Nawaz - implementation of various heuristic algorithms including HUIM-HC, HUIM-GA, LCIM and MaxFEM
  • Yimin Zhang - implementation of the FCHM_bond and FCHM_allconfidence algorithms with P. Fournier-Viger, implementations of the LHUI-Miner and PHUI-Miner algorithms, and provided several datasets with timestamps.
  • Ying Wang - implementation of TSPIN and MLHUI-Miner
  • Azadeh Soltani - implementations of the following algorithms: MISApriori, CFPGrowth++, estDec
  • Xuan Liu et al. (github liuxuan615) - code of the ETAUIM, EMSFUI-D, EMSFUI-B, FUIMTF-Tree and FUIMTWU-Tree algorithms (obtained under the GPL license as it contain GPL code)
  • Prashant Barhate - implementations of the IHUP, UP-Growth and UPGrowth+ algorithms
  • Chao Cheng - implementations of gSPAN, TSeqMiner and TKG algorithms.
  • Vincent Nofong - implementations of the PPFP, SRPFPM and NPFPM algorithms.
  • Jiaxuan Li - implementations of the CEPB, corCEPB, CEPN algorithms
  • Yuechun Li - implemation of VertTIRP, FastTIRP, ImpTIRP, etc.
  • Yanjun Yang - implemation of LTHUI-Miner and TKTE
  • Ganghuan He - implementation of AER-Miner
  • Mourad Nouioua - implementation of TKC, FQHUI-Miner, TKQ, CHUQI-Miner ...
  • Farid Nouioua and Oualid Ouarem - implementation of NONEPI and others...
  • Nader Aryabarzan - implementations of the dFIN, Neg-FIN, NEclatClosed, NAFCP algorithms
  • Souleymane Zida - implementation of the EFIM and HUSRM algorithms with P. Fournier-Viger
  • Zhihong Deng - implementations of the PrePost, PrePost+ and FIN algorithms
  • Fabio Fumarola, Pasqua Fabiana Lanotte, Michelangelo Ceci, Donato Malerba, Eliana Salvemini, Jiawei Han - implementations of the FAST and CloFAST algorithms
  • Zhitian Li - implementations of the MPFPS-DFS and MPFPS-BFS algorithms.
  • Bay Vo, Loan T. T. Nguyen, N.T. Tung, and colleagues - implementation of CLH-Miner and FEACP.
  • Muhang Li et al. (github: limuhangk)- implementation of CLS-Miner and HMiner-Closed algorithms, obtained under the GPL license from Github.
  • Fang et al. (github: jnfrancis)- implementation of HUIM-SU algorithm, obtained under the GPL license from Github.
  • Xiang Li - translation of Opus-Miner to Java from original GPL code of G. Webb
  • Hoang Thanh Lam, Toon Calders, Fabian Moerchen, Dmitriy Fradkin - implementations of the GoKrimp and SeqKrimp algorithms
  • Alan Souza - implementations of the LCM algorithms
  • Sabarish Raghu - implementation of the text clusterer, and document classifier
  • Ahmed El-Serafy, Hazem El-Raffiee - implementation of the GCD algorithm
  • Ryan Panos - implementation of a version of CMDeo with the lift measure
  • F. Padillo. J. Maria Luna, S. Ventura - implementations of several association rule classifiers such as CBA, CBA2 and, ACAC was obtained from their LAC project under the GPL license, and then adapted..
    Yuriy Guskov - the time series viewer of SPMF includes some code from the simple java plot viewer made by Y. Guskov
  • Petijean et al. - code of Skopus, obtained under the GPL license
  • Peng et al. the code of mHUIMiner under GPL license obtained from the Github repository.
  • Sacha Servan-Schreiber - implementation of the Prosecco algorithm
  • Pushp, Satish Chand - implementation of the R-Miner algorithm
  • Hoang Thi Dieu - improvements to the FHSAR implementation
  • Blake Johns, Ryan Benton - implementation of the RPGrowth algorithm
  • Lens Feremans, Cule Boris, Bart Goethals - original implementation of the QCSP algorithm
  • Mickael Da-Silva - modification of the Fast algorithm to add the maximum support constraint
  • Zevin Shaul - implementation of the cgSpan algorithm, optimizations to TopKRules and other algorithms.
  • Srikumar Krishnamoorthy - implementation of the THUI algorithm
  • Lai et al. - implementation of the PHMN and PHMN+ (obtained under GPL license from Github)
  • aman955 - implementation of the ECHUM algorithm (obtained under the GPL license from Github)
  • Jayakrushna Sahoo - implementation of the FHIM and HUCI-Miner algorithms for high utility association rules

Code optimizations

  • Dan Cappucio - an important optimization of the FPGrowth implementation

User interface

  • Hanane Amirat - provided feedback to improve the user interface design

Performance evaluation

  • Rincy N. Thomas - performance comparison of sequential pattern mining algorithms on various datasets, and found several errors on the website.

Textbooks

Datasets

  • Zhongjie Zhang - provided several datasets with item labels in SPMF format, converted from the UCI repository (Skin, USCensus, PAMP, OnlineRetail, RecordLink, PowerC, SUSY, kddcup99, (Chicago)
  • Ashwin Balani - provided MatLab code for dataset generation (available on the "datasets" page)
  • Saqib Nawaz - provided the ProofSequences dataset
  • Dang Nguyen - provided several frequent subgraph mining datasets
  • Jiaxuan Li - provided datasets for cost/utility sequence mining
  • Frederic Flouvat - prepared several datasets for sequential pattern mining
  • Mourad Nouioua - prepared datasets for high utility quantitative itemset mining

Installation and command line interface

  • Antonio Sergio Ando - feedback and some code for the command line interface, bug fix, and ANT script (to be included).

Bug reports,  bug fixes and other contributions

  • Ludwig Zellner and Gabriel Lindner - fixed a bug of the TKG algorithm when using the user interface, and reported another bug related to the "output single vertice" option.., fixed a bug in the output of CMRules
  • Huan Yang - Fixed a bug related to input file reading in the TNR algorithm.
  • Jose Maria Luna - reported a bug for TKO when called from the graphical user interface.
  • Zefen Chen - reported a but in the NONEPI algorithm
  • Victor Rodriguez-Fernandez - reported bugs related to GoKrimp
  • Devin Shaul - reported bugs related to TKG and gSpan
  • Lili Chen - reported bug in HUSRM
  • Bao Vu - fixed a bug in the CHUI-Miner(max) implementation
  • Chaomin Huang - fixed deep code conversion bugs of the HUIM-GA.HUIM-GA-tree, HUIM-BPSO-tree, and HUIM-BPSO algorithms
  • Hung Nguyen - fixed a bug in the MinFHM algorithm
  • Martin Fekete - reported a bug about resetting the memory statistics in BIDE and PrefixSpan
  • Matthieu Gousseff - reported a bug related to sequence identifiers in sequential pattern mining algorithms, and errors in the documentation
  • Luis Angerstein and Jan Wolter fixed a few bugs in the TDAG and LZ78 sequence prediction models
  • Konstantin Bottcher - reported a bug in FPGrowth and unused variables and condition in FPGrowth to improve the code
  • C. Sivamathi - reported a bug in TS-HOUN x
  • Rashmie Abeysinghe - reported a bug in AlgoArrays.Java
  • Benjamin Andow - reported a bug in the generation of closed association rule mining with FPClose algorithm
  • Bima Haryanto Putra - reported a bug in the TopKRules algorithm
  • Muhammad Yasir Chaudhry - reported a bug in the Apriori algorithm
  • with length constraint
  • Srikumar Krishnamoorthy - reported a problem in some utility mining datasets
  • Majdi Mafarja - reported a bug in the HUIM-bso algorithm
  • Tai Dinh - reported a bug in the USpan algorithm
  • Tin Truong Chi - reported a bug in the USpan algorithm
  • Tarannum Zaman - reported a bug in closed association rule mining with FPClose
  • Natalia Mord - reported a bug in the MaxSP algorithm
  • Andrey Shestakov - reported a bug in the command line interface
  • Antoine Pigeau - reported a bug in the VMSP algorithm
  • Himel Dev - reported a bug in the VMSP algorithm
  • Slimane Oulad Naoui - suggested to better handle incorrect algorithm parameters
  • Tin Truong Chi - fixed a bug in the ClaSP and CM-Clasp algorithms
  • Yimin Zhang - reported a problem in the FOSHU and TS-HOUN algorithms
  • Gehad Ahmed Soltan Abd-Elaleem - reported a bug in the FHSAR
  • algorithm
  • Preethy Varma - reported a bug in ClaSP and CM-Clasp algorithms
  • Mike Rostermund - reported a problem in the output of SPAM based algorithms
  • Jaroslav Fowkes and Thomas Christie - reported a bug in the GoKrimp implementation
  • Jamshi Nazeer - reported a bug in the FPClose implementation
  • Insu Yun - reported a bug in the FPClose implementation
  • Dharmen Punjani - reported a bug in the text clusterer
  • Martin Bockle - reported a bug in the GUI of SPMF
  • Pierre-Emmanuel Leroy - reported a bug in the Cori algorithm
  • Masanori Akiyoshi- reported a bug in the FPGrowth algorithm
  • Asmaa - reported a bug in the Zart algorithm
  • Acquah Hackman - reported a bug in the LHUI-Miner algorithm
  • Choong Shin Siang and Wong Li Pei - reported a bug on the use of the "maxgap" constraint
  • for SPAM-based algorithms.
  • Ryan G. Benton - reported and fixed a bug in the Itemset-Tree and Memory Efficient Itemset-Tree
  • G. Gutierrez - reported a bug about how ID3 trees are printed to console
  • Mehran Memon - reported a bug in the BIDE+ implementation
  • Nahumi - reported a bug for running the "Fournier08-Closed+time" algorithm using the GUI
  • Wen Zhang - reported bugs: a bug that was generating headless exceptions, and a bug about command line arguments
  • Abdalghani Abujabal - reported bugs/improvement for the ECLAT
  • algorithm
  • C. Albert Thompson - reported a bug in association rule generation with CFPGrowth, and a bug in TNS/TopSeqRules
  • Manperta Negara Situmorang - reported a bug in association rule generation with CFPGrowth
  • Michael Witbrock - reported an issue with character encoding of source code files
  • Cheng Zhou - reported inconsistencies in sequential pattern mining algorithms source code
  • Vathsala.H - reported a bug in the ID3 implementation
  • Rai. A. - reported a bug in using the hierarchical clustering algorithm with the GUI
  • Arina Pramudita - reported some minsup rounding inconsistencies between some sequential pattern mining algorithms
  • Radhika Loombas - reported duplicated variables and unreachable code in FPGrowth
  • Srinivas K. - bug report for the sequence database generator and for the Charm algorithm
  • Faisal Feroz - bug report for the AbstractOrderedItemset and ItemsetTree classes
  • Peter Toth - bug report for association rule mining with FPGrowth, and bug report for TNS / TopSeqRules
  • saiph..@... - reported a bug in the GUI
  • Antonio Sergio Ando - reported a bug in FHSAR
  • Said Hamani - reported a bug in the INDIRECT
  • algorithm
  • Faezeh Jafari - reported a bug in Cluster.java
  • Dvijesh Bhatt - reported a bug in SPAM
  • shouwangji@... reported unused variables and other minor problems in PrefixSpan
  • Brock - reported a bug in BIDE +
  • G. Bruno - reported bugs in BIDE+ and FPGrowth
  • E. Schubert - reported a bug in DBScan
  • A. Pardeshi - reported a bug in ECLAT / CHARM
  • Abdel Naser Pouamoun - reported a bug in the Cluster class
  • Minh Pham - reported a bug in ERMiner
  • Jiaxing Mai - reported various errors in the documentation and code.
  • Darrell Conklin - reported a bug in VMSP and VGEN related to maxgap
  • Lucas Potin - reported errors in the dataset page of the SPMF website.
  • Alexandre Vernotte - reported a bug in VMSP related to maxgap
  • Suzuki Shota - reported a bug about sequential pattern mining with the Online_retail dataset.
  • M. Zivanovic - reported a bug in NOSEP

Antonio Gomariz Penalver and Philippe Fournier-Viger