Sequential Pattern Mining

These algorithms discover sequential patterns in a set of sequences. For a good overview of sequential pattern mining algorithms, please read this survey paper.

• algorithms for mining sequential patterns (subsequences that appear in many sequences) of a sequence database
  • the CM-SPADE algorithm (Fournier-Viger et al., 2014, 📊 slides)
  • the CM-SPAM algorithm (Fournier-Viger et al., 2014, 📊 slides)
  • the FAST algorithm (Salvemini et al., 2011)
  • the GSP algorithm (Srikant et al., 1996)
  • the LAPIN (aka LAPIN-SPAM) algorithm (Yang et al., 2005)
  • the PrefixSpan algorithm (Pei et al., 2004, 📊 slides ▶ video)
  • the SPADE algorithm (Zaki et al., 2001)
  • the SPAM algorithm (Ayres et al., 2002)
• algorithms for mining closed sequential patterns in a sequence database
  • the ClaSP algorithm (Gomariz et al., 2013)
  • the CM-ClaSP algorithm (Fournier-Viger et al., 2014, 📊 slides)
  • the CloFAST algorithm (Fumarola et al., 2016)
  • the CloSpan algorithm (Yan et al., 2003)
  • the BIDE+ algorithm (Wang et al., 2007)
• algorithms for mining maximal sequential patterns in a sequence database
  • the VMSP algorithm (Fournier-Viger et al., 2014, 📊 slides)
  • the MaxSP algorithm (Fournier-Viger et al., 2013, 📊 slides)
• algorithms for mining the top-k sequential patterns in a sequence database
  • the TKS algorithm (Fournier-Viger et al., 2013, 📊 slides)
  • the TSP algorithm (Tzvetkoz et al., 2003)
  • the Skopus algorithm – top-k sequential patterns using leverage and significance (Petijean et al., 2016)
• algorithms for mining sequential generator patterns in a sequence database
  • the VGEN algorithm (Fournier-Viger et al., 2014)
  • the FEAT algorithm (Gao et al., 2008)
  • the FSGP algorithm (Yi et al., 2011)
• algorithms for mining nonoverlapping sequential patterns in one or many sequences of symbols/characters (can count multiple occurrences per sequence)
  • the NOSEP algorithm (Wu et al., 2018) new
• algorithms for mining compressing sequential patterns
  • the GoKrimp and SeqKrimp algorithms (Lam et al., 2012; Lam et al., 2014)
  • the HMG-GA and HMG-SA algorithms – compressing patterns in genome sequences (Nawaz et al., 2025) new
• algorithm for identifying the top-k quantile based cohesive sequential patterns in a single sequence or in multiple sequences
  • the QCSP algorithm (Feremans et al., 2019)
• algorithms for mining multidimensional sequential patterns in a multidimensional sequence database
  • the SeqDIM algorithm – frequent multidimensional sequential patterns (Pinto et al., 2001)
  • the Songram et al. algorithm – frequent closed multidimensional sequential patterns (Songram et al., 2006)
• the Fournier-Viger et al. algorithm – a sequential pattern mining algorithm combining features from well-known algorithms with original extensions (Fournier-Viger et al., 2008):
  • mining sequences with minimum support by database-projection (based on PrefixSpan, Pei et al., 2004)
  • mining sequences with min/max time interval and sequence time length (based on Hirate-Yamana, 2006)
  • mining closed sequences (based on BIDE+, Wang et al., 2007)
  • mining multi-dimensional sequences (based on Pinto et al., 2001)
  • mining closed multi-dimensional sequences (based on Songram et al., 2006 and Pasquier et al., 1999)
  • mining sequences with integer-valued items and automatic clustering of values (Fournier-Viger et al., 2008)
• algorithm for mining high-utility sequential patterns in a sequence database
  • the USPAN algorithm (Yin et al., 2012)
• algorithm for mining cost-efficient sequential patterns (a.k.a. low-cost high utility sequential patterns)
  • the CorCEPB algorithm – binary utility with cost values (Fournier-Viger et al., 2020, 📊 slides ▶ video) new
  • the CEPB algorithm – binary utility with cost values (positive utility only) (Fournier-Viger et al., 2020, 📊 slides ▶ video) new
  • the CEPN algorithm – numeric utility with cost values (Fournier-Viger et al., 2020, 📊 slides ▶ video) new
• algorithm for mining high-utility probability sequential patterns in a sequence database
  • the PHUSPM algorithm (Zhang et al., 2018)
  • the UHUSPM algorithm (Zhang et al., 2018)
• algorithm for progressive sequential pattern mining with convergence guarantees
  • the ProSecCo algorithm (Servan-Schreiber et al., 2018)
• algorithms for mining sequential patterns with flexible constraints in a time-extended sequence database (e.g. MOOC data)
  • the SPM-FC-L algorithm (Song et al., 2022) new
  • the SPM-FC-P algorithm (Song et al., 2022) new
• the Occur algorithm – finds all occurrences of sequential patterns in sequences by post-processing
• algorithms for mining patterns in sequences of events described by time intervals (Time Interval Related Pattern — TIRP — mining)
  • the FastTIRP algorithm (Fournier-Viger et al., 2022) new
  • the VertTIRP algorithm (Mordvanyuk et al., 2021) new

Sequential Rule Mining

These algorithms discover sequential rules in a set of sequences.

• algorithms for mining sequential rules in a sequence database
  • the ERMiner algorithm (Fournier-Viger et al., 2014)
  • the RuleGrowth algorithm (Fournier-Viger et al., 2011, Fournier-Viger et al., 2015, 📊 slides ▶ video)
  • the CMRules algorithm (Fournier-Viger et al., 2010, 📊 slides ▶ video)
  • the CMDeo algorithm (Fournier-Viger et al., 2010)
  • the RuleGen algorithm (Zaki et al., 2001)
• algorithms for mining sequential rules in a sequence database with the window size constraint
  • the TRuleGrowth algorithm (Fournier-Viger, 2012, Fournier-Viger et al., 2015)
• algorithms for mining top-k sequential rules in a sequence database
  • the TopSeqRules algorithm (Fournier-Viger et al., 2011, 📊 slides)
  • the TopSeqClassRules algorithm – top-k class sequential rules (variation of Fournier-Viger et al., 2011)
  • the TNS algorithm – top-k non-redundant sequential rules (Fournier-Viger, 2013)
• algorithm for mining high-utility sequential rules in a sequence database
  • the HUSRM algorithm (Zida et al., 2015)

Sequence Prediction

These algorithms predict the next symbol of a sequence based on a set of training sequences.

• the Compact Prediction Tree+ (CPT+) algorithm (Gueniche et al., 2015, 📊 slides ▶ video)
• the Compact Prediction Tree (CPT) algorithm (Gueniche et al., 2013, ▶ video)
• the First-order Markov Chains (PPM – order 1) (Clearly et al., 1984)
• the Dependency Graph (DG) (Padmanabhan, 1996)
• the All-k-Order Markov Chains (AKOM) (Pitkow, 1999)
• the TDAG (Laird & Saul, 1994)
• the LZ78 (Ziv, 1978)

Itemset Mining

These algorithms discover interesting itemsets (sets of values) in a transaction database. For a good overview, please read this survey paper.

• algorithms for discovering frequent itemsets in a transaction database
  • the Apriori algorithm (Agrawal & Srikant, 1994, ▶ video) new
  • the AprioriTID algorithm (Agrawal & Srikant, 1994)
  • the FP-Growth algorithm (Han et al., 2004)
  • the Eclat algorithm (Zaki, 2000, ▶ video)
  • the dEclat algorithm (Zaki and Gouda, 2001, 2003)
  • the Relim algorithm (Borgelt, 2005)
  • the H-Mine algorithm (Pei et al., 2007)
  • the LCMFreq algorithm (Uno et al., 2004)
  • the PrePost algorithm (Deng et al., 2012)
  • the PrePost+ algorithm (Deng & Lv, 2015)
  • the FIN algorithm (Deng et al., 2014)
  • the DFIN algorithm (Deng et al., 2016)
  • the NegFIN algorithm (Aryabarzan et al., 2018)
  • the DIC algorithm (Brin et al., 1997) new
  • the TM algorithm (Song et al., 2006) new
  • the SAM algorithm (Borgelt et al., 2009) new
  • the LinearTable algorithm (Lu et al., 2023) new
• algorithms for discovering frequent closed itemsets in a transaction database
  • the FPClose algorithm (Grahne and Zhu, 2005)
  • the Charm algorithm (Zaki and Hsiao, 2002)
  • the dCharm algorithm (Zaki and Gouda, 2001)
  • the DCI_Closed algorithm (Lucchese et al., 2004)
  • the LCM algorithm (Uno et al., 2004)
  • the AprioriClose (aka Close) algorithm (Pasquier et al., 1999)
  • the AprioriTID Close algorithm (Pasquier et al., 1999, Agrawal & Srikant, 1994)
  • the NAFCP algorithm (Le et al., 2015)
  • the NEclatClosed algorithm (Aryabarzan et al., 2021)
  • the CARPENTER algorithm (Pan et al., 2003) new
  • the DBVMiner algorithm (Vo et al., 2012) new
  • the NEWCHARM algorithm (Ye et al., 2015) new
• algorithms for recovering all frequent itemsets from frequent closed itemsets
  • the LevelWise algorithm (Pasquier et al., 1999)
  • the DFI-Growth algorithm (Huang et al., 2019)
  • the DFI-List algorithm (Wu et al., 2020)
• algorithms for discovering frequent maximal itemsets in a transaction database
  • the FPMax algorithm (Grahne and Zhu, 2003)
  • the Charm-MFI algorithm (Szathmary et al., 2006)
  • the CARPENTER-MAX algorithm (Pan et al., 2003) new
  • the GENMAX algorithm (Gouda et al., 2005) new
• algorithms for mining frequent itemsets with multiple minimum supports
  • the MSApriori algorithm (Liu et al., 1999)
  • the CFPGrowth++ algorithm (Uday & Reddy, 2011, Hu & Chen, 2006)
• algorithms for mining generator itemsets in a transaction database
  • the DefMe algorithm (Soulet & Rioult, 2014)
  • the Talky-G and Talky-G-Diffset algorithms (Szathmary et al., 2009) new
  • the Pascal algorithm (Bastide et al., 2002)
  • the Zart algorithm (Szathmary et al., 2007)
• algorithms for mining rare itemsets and/or correlated itemsets in a transaction database
  • the AprioriInverse algorithm – perfectly rare itemsets (Koh & Roundtree, 2005, ▶ video) and AprioriTIDInverse (vertical structure variant)
  • the AprioriRare algorithm – minimal rare itemsets and frequent itemsets (Szathmary et al., 2007, ▶ video) and AprioriTIDRare (vertical structure variant)
  • the CORI algorithm – minimal rare correlated itemsets using support and bond measures (Bouasker et al., 2015, ▶ video)
  • the RP-Growth algorithm (Tsang et al., 2011) new
• algorithms for performing targeted and dynamic queries about association rules and frequent itemsets
  • the Itemset-Tree – incrementally updatable data structure with query algorithms (Kubat et al., 2003)
  • the Memory-Efficient Itemset-Tree (Fournier-Viger, 2013, 📊 slides)
• algorithms to discover frequent itemsets in a stream
  • the estDec algorithm – recent frequent itemsets (Chang & Lee, 2003)
  • the estDec+ algorithm – recent frequent itemsets (Shin et al., 2014)
  • the CloStream algorithm – frequent closed itemsets (Yen et al., 2009)
• the U-Apriori algorithm – frequent itemsets in uncertain data (Chui et al., 2007)
• the VME algorithm – erasable itemsets (Deng & Xu, 2010)
• algorithms to discover fuzzy frequent itemsets in a quantitative transaction database
  • the FFI-Miner algorithm (Lin et al., 2015)
  • the MFFI-Miner algorithm (Lin et al., 2016)
• the OPUS-Miner algorithm – self-sufficient itemsets (Webb et al., 2014)
• algorithms to discover compressing itemsets based on the MDL principle
  • the KRIMP algorithm (Vreeken et al., 2011) new
  • the SLIM algorithm (Smets et al., 2012) new
  • the GRIMP algorithm (Nawaz et al., 2025) new
  • the HMP-SA algorithm (Chen et al., 2026) new
  • the HMP-HC algorithm (Chen et al., 2026) new
• algorithms to discover the top-k most frequent itemsets
  • the Apriori(top-k) algorithm (modified Apriori) new
  • the FPGrowth(top-k) algorithm (modified FP-Growth) new

Episode Mining

These algorithms discover patterns (episodes) in a single sequence of events. For a good overview, please read this survey paper.

• algorithms for mining frequent episodes
  • the EMMA algorithm – frequent episodes based on head frequency (Kuo-Yu et al., 2008)
  • the AFEM algorithm – frequent episodes based on head frequency (Fournier-Viger et al., 2022) new
  • the MINEPI+ algorithm – frequent episodes based on head frequency (Kuo-Yu et al., 2008)
  • the MINEPI algorithm – frequent episodes, minimal occurrences, no simultaneous events (Mannila & Toivonen, 1997)
  • the TKE algorithm – top-k most frequent episodes (Fournier-Viger et al., 2020)
  • the MaxFEM algorithm – maximal frequent episodes (Fournier-Viger et al., 2022) new
  • the EMDO algorithm – frequent parallel episodes based on distinct occurrences (Ouarem et al., 2024) new
• algorithms for mining episode rules
  • the POERM algorithm – partially-ordered episode rules, non-overlapping support (Fournier-Viger et al., 2021, ▶ video 📊 slides) new
  • the POERM-ALL algorithm – all partially-ordered episode rules (Fournier-Viger et al., 2021, ▶ video 📊 slides) new
  • the POERMH algorithm – partially-ordered episode rules, head support (Fournier-Viger et al., 2021, ▶ video 📊 slides) new
  • the NONEPI algorithm – episode rules using non-overlapping frequency (Ouarem et al., 2021)
  • the EMDO-Rules and EMDOP-Rules algorithms – episode rules from parallel episodes (Ouarem et al., 2024) new
  • algorithms to generate episode rules (Mannila & Toivonen, 1997) using output of TKE, AFEM, EMMA or MINEPI+ new
• algorithms for mining high-utility episodes in a sequence of complex events with utility information
  • the HUE-SPAN algorithm (Fournier-Viger et al., 2019, 📊 slides) new
  • the US-SPAN algorithm (Wu et al., 2013)
  • the TUP algorithm – top-k high-utility episodes (Rathore et al., 2016)
• algorithms for mining nonoverlapping sequential patterns in one or many sequences of symbols
  • the NOSEP algorithm (Wu et al., 2018) new
• algorithms for mining frequent sequential patterns with periodic wildcard gaps in a sequence of characters
  • the MAPD algorithm (Wu, Y. et al., 2014)
• algorithms for mining self-adaptive one-off weak-gap strong sequential patterns in a sequence of characters
  • the OWSP-Miner algorithm (Wu, Y. et al., 2022) new

Periodic Pattern Mining

These algorithms discover patterns that periodically appear in a sequence of records (e.g. transactions).

• algorithms for finding frequent periodic patterns in a single sequence of events
  • the PFPM algorithm (Fournier-Viger et al., 2016, 📊 slides ▶ video)
• algorithms for mining stable periodic itemsets in a sequence of events with or without timestamps
  • the SPP-Growth algorithm (Fournier-Viger et al., 2019, 📊 slides ▶ video)
  • the TSPIN algorithm – top-k stable periodic frequent itemsets (Fournier-Viger et al., 2021) new
• algorithms for mining locally periodic patterns in a transaction database with or without timestamps
  • the LPP-Growth algorithm (Fournier-Viger, 2020, 📊 slides) new
  • the LPPM_breadth algorithm (Fournier-Viger, 2020, 📊 slides) new
  • the LPPM_depth algorithm (Fournier-Viger, 2020, 📊 slides) new
• algorithms for discovering periodic patterns that are significant or non-redundant
  • the NPFPM algorithm – non-redundant periodic frequent itemsets (Afriyie et al., 2020, 2021) new
  • the PPFP algorithm – productive periodic frequent itemsets (Nofong, 2016) new
  • the SRPFPM algorithm – self-reliant periodic frequent patterns (Nofong et al., 2021) new
• algorithms for mining periodic high-utility itemsets in a sequence of transactions with utility information
  • the PHM algorithm (Fournier-Viger et al., 2016, 📊 slides ▶ video)
  • the PHMN algorithm (2023) – periodic high-utility itemsets with positive or negative utility new
  • the PHMN+ algorithm (2023) – periodic high-utility itemsets with positive or negative utility new
  • the PHM_irregular algorithm – irregular (non-periodic) high-utility itemsets (variation of PHM)
• algorithms for finding periodic patterns in multiple sequences of events
  • the MPFPS_BFS algorithm (Fournier-Viger et al., 2019, 📊 slides)
  • the MPFPS_DFS algorithm (Fournier-Viger et al., 2019, 📊 slides)
• algorithms for mining rare correlated periodic patterns common to multiple sequences
  • the MRCPPS algorithm (Fournier-Viger et al., 2020) new

Graph Pattern Mining new

These algorithms discover patterns in graphs.

• algorithms for mining frequent subgraphs
  • the TKG algorithm – top-k frequent subgraphs (Fournier-Viger, 2019, 📊 slides)
  • the gSpan algorithm – all frequent subgraphs (Yan et al., 2002)
  • the cgSpan algorithm – frequent closed subgraphs (graph database or single graph; traditional or MNI support) (Shaul et al., 2021) new
• algorithms for mining patterns in a dynamic attributed graph
  • the TSeqMiner algorithm (Fournier-Viger et al., 2019)
  • the AER-Miner algorithm (Fournier-Viger et al., 2020, 📊 slides)

High-Utility Pattern Mining

These algorithms discover patterns having a high utility (importance) in different kinds of data. For a good overview, read the survey paper or the high utility-pattern mining book.

• algorithms for mining high-utility itemsets in a transaction database having profit information
  • the EFIM algorithm (Zida et al., 2016, 📊 slides)
  • the FHM algorithm (Fournier-Viger et al., 2014, 📊 slides ▶ video)
  • the HUI-Miner algorithm (Liu & Qu, 2012, ▶ video)
  • the HUP-Miner algorithm (Krishnamoorthy, 2014)
  • the mHUIMiner algorithm (Peng et al., 2017)
  • the UFH algorithm (Dawar et al., 2017)
  • the HMiner algorithm (Krishnamoorthy, 2017)
  • the ULB-Miner algorithm (Duong et al., 2018) new
  • the IHUP algorithm (Ahmed et al., 2009)
  • the Two-Phase algorithm (Liu et al., 2005)
  • the UP-Growth algorithm (Tseng et al., 2011)
  • the UP-Growth+ algorithm (Tseng et al., 2013)
  • the UP-Hist algorithm (Dawar et al., 2015)
  • the d2HUP algorithm (Liu et al., 2012)
  • the FHIM algorithm (Sahoo et al., 2015)
  • the PUCPMiner algorithm (Patel et al., 2022) new
  • the RMiner algorithm (Sra et al., 2023) new
• algorithm for efficiently mining high-utility itemsets with length constraints
  • the FHM+ algorithm (Fournier-Viger et al., 2016, 📊 slides)
• algorithm for mining correlated high-utility itemsets
  • the FCHM_bond algorithm – bond measure (Fournier-Viger et al., 2016, 📊 slides ▶ video)
  • the FCHM_allconfidence algorithm – all-confidence measure (Fournier-Viger et al., 2016, 📊 slides)
  • the ECHUM algorithm (Ramesh et al., 2022) - obtained from Github aman955 under the GPL license
• algorithm for mining high-utility itemsets with negative unit profit values
  • the FHN algorithm (Fournier-Viger et al., 2014, 📊 slides)
  • the HUINIV-Mine algorithm (Chu et al., 2009)
• algorithms for mining multi-level or cross-level high-utility itemsets with a taxonomy
  • the CLH-Miner (Fournier-Viger et al., 2020, 📊 slides)
  • the FEACP (Tung et al., 2022) new
  • the MLHUI-Miner (Cagliero et al., 2017)
• algorithm for mining low-cost high-utility itemsets
  • the LCIM algorithm (Fournier-Viger et al., 2022) new
• algorithm for mining frequent high-utility itemsets
  • the FHMFreq algorithm (variation of FHM, Fournier-Viger et al., 2014)
• algorithm for mining on-shelf high-utility itemsets (items with time periods)
  • the FOSHU algorithm (Fournier-Viger et al., 2015, 📊 slides)
  • the TS-HOUN algorithm (Lan et al., 2014)
• algorithm for incremental high-utility itemset mining
  • the EIHI algorithm (Fournier-Viger et al., 2015, 📊 slides)
  • the HUI-LIST-INS algorithm (Lin et al., 2014)
• algorithm for incremental closed high-utility itemset mining
  • the IncCHUI algorithm (Dam et al., 2018) - obtained from Github under the GPL license
• algorithm for mining concise representations of high-utility itemsets
  • the HUG-Miner – high-utility generators (Fournier-Viger et al., 2014, 📊 slides)
  • the GHUI-Miner – generators of high-utility itemsets (Fournier-Viger et al., 2014, 📊 slides)
  • the MinFHM – minimal high-utility itemsets (Fournier-Viger et al., 2016, 📊 slides ▶ video)
  • the EFIM-Closed – closed high-utility itemsets (Fournier-Viger et al., 2016, 📊 slides)
  • the CHUI-Miner – closed high-utility itemsets (Wu et al., 2015)
  • the CLS-Miner – closed high-utility itemsets (Dam et al., 2019) new
  • the HMiner_Closed – closed high-utility itemsets (Nguyen et al., 2019) new
  • the CHUD – closed high-utility itemsets (Tseng et al., 2015)
  • the CHUI-Miner(Max) – maximal high-utility itemsets (Wu et al., 2019) new
  • the HUCI_Miner – closed high-utility itemsets and generators (Sahoo et al., 2015) new
• algorithm for mining the skyline high-utility itemsets
  • the SkyMine algorithm (Goyal et al., 2015)
  • the SFUI_UF algorithm – skyline frequent high-utility itemsets (Song et al., 2021) new
  • the SFU_CE algorithm – skyline frequent high-utility itemsets (Song et al., 2021, 📊 slides) new
  • the SFUPMinerUemax algorithm (Lin et al., 2016)
  • the EMSFUI_D algorithm (Liu et al., 2022) new
  • the EMSFUI_B algorithm (Liu et al., 2022) new
• algorithm for mining the top-k high-utility itemsets
  • the TKU algorithm (Tseng et al., 2015) - obtained from the UP-Miner under the GPL license
  • the TKO-Basic algorithm (Tseng et al., 2015)
  • the THUI algorithm (Krishnamoorthy, 2019) new
• algorithms for mining the top-k high-utility itemsets from a data stream
  • the FHMDS and FHMDS-Naive algorithms (Dawar et al., 2017)
• algorithm for mining quantitative high-utility itemsets
  • the FHUQI-Miner algorithm (Nouioua et al., 2021, 📊 slides) new
  • the VHUQI algorithm (Wu et al., 2014)
  • the TKQ algorithm – top-k patterns (Nouioua et al., 2021, 📊 slides ▶ video) new
  • the CHUQI-Miner – correlated quantitative high-utility itemsets (Nouioua et al., 2021, 📊 slides ▶ video) new
• algorithm for mining high-utility sequential rules in a sequence database
  • the HUSRM algorithm (Zida et al., 2015)
• algorithm for mining high-utility sequential patterns in a sequence database
  • the USPAN algorithm (Yin et al., 2012)
• algorithm for mining cost-efficient sequential patterns (low-cost high-utility)
  • the CorCEPB algorithm - mine patterns in sequences with binary utility information and cost values (Fournier-Viger et al., 2020, 📊 slides ▶ video) new
  • the CEPB algorithm - mine patterns in sequences with binary utility information and cost values - consider only sequence with positive utility (Fournier-Viger et al., 2020, 📊 slides) new
  • the CEPN algorithm - mine patterns  in sequences with numeric utility information and cost values (Fournier-Viger et al., 2020, 📊 slides ▶ video) new
• algorithm for mining high-utility probability sequential patterns
  • the PHUSPM algorithm (Zhang et al., 2018)
  • the UHUSPM algorithm (Zhang et al., 2018)
• algorithm for heuristically mining the top-k high-utility itemsets
  • the TKU-CE algorithm (Song et al., 2021) new
  • the TKU-CE+ algorithm (Song et al., 2021) new
• algorithm for mining high-utility itemsets using evolutionary algorithms, swarm intelligence or other meta-heuristics
  • the HUIM-AF algorithm (Song et al., 2021) new
  • the HUIM-HC algorithm (Fournier-Viger et al., 2021) new
  • the HUIM-SA algorithm (Fournier-Viger et al., 2021) new
  • the HUIM-ACO algorithm (Song et al., 2020) new
  • the HUIM-SPSO algorithm (Song et al., 2020) new
  • the HUIF-PSO algorithm (Song et al., 2018)
  • the HUIF-GA algorithm (Song et al., 2018)
  • the HUIF-BA algorithm (Song et al., 2018)
  • the HUIM-ABC algorithm (Song et al., 2018)
  • the HUIM-GA algorithm (Kannimuthu et al., 2014)
  • the HUIM-BPSO algorithm (Lin et al., 2016)
  • the HUIM-GA-tree algorithm (Lin et al., 2016)
  • the HUIM-BPSO-tree algorithm (Lin et al., 2016)
• algorithm for mining high average-utility itemsets
  • the HAUI-Miner (Lin et al., 2016)
  • the EHAUPM (Lin et al., 2017)
  • the HAUIM-GMU (Song et al., 2021) new
  • the HAUI-MMAU – with multiple thresholds (Lin et al., 2016)
  • the MEMU – with multiple thresholds (Lin et al., 2018)
• algorithms for mining the top-k high average-utility itemsets
  • the ETAUIM algorithm (2023) new - obtained under the GPL license from Github liuxuan615 
• algorithms for mining high-utility episodes in a sequence of complex events containing utility information
  • the HUE-SPAN algorithm (Fournier-Viger et al., 2019, 📊 slides) new
  • the TUP algorithm (Rathore et al., 2016)
  • the UP-SPAN algorithm (Wu et al., 2013)
• algorithms for mining periodic high-utility patterns
  • the PHM algorithm (Fournier-Viger et al., 2016, 📊 slides)
  • the PHMN and PHMN+ algorithms (2023) - obtained from Github @laughing1999 under the GPL license)
• algorithms for discovering irregular high-utility itemsets
  • the PHM_irregular algorithm (variation of PHM) new
• algorithm for discovering local high-utility itemsets
  • the LHUI-Miner (Fournier-Viger et al., 2019, 📊 slides) new
• algorithm for discovering peak high-utility itemsets
  • the PHUI-Miner (Fournier-Viger et al., 2019, 📊 slides) new
• algorithm for discovering locally trending high-utility itemsets
  • the LTHUI-Miner (Fournier-Viger et al., 2020, ▶ video 📊 slides) new
• algorithm for high-utility itemset mining with a recency constraint
  • the ScentedUtilityMiner algorithm (Sra et al., 2024) new
• algorithm for discovering high-utility association rules
  • the HGB_all algorithm – all or non-redundant high-utility association rules (Sahoo et al., 2015) new
  • the HGB algorithm – non-redundant high-utility association rules (Sahoo et al., 2015) new

Association Rule Mining

These algorithms discover interesting associations between symbols (values) in a transaction database (database records with binary attributes).

• an algorithm for mining all association rules with the confidence measure (Agrawal & Srikant, 1994, ▶ video)
• an algorithm for mining all association rules with the lift measure (adapted from Agrawal & Srikant, 1994)
• an algorithm for mining the IGB informative and generic basis of association rules (Gasmi et al., 2005)
• an algorithm for mining perfectly sporadic association rules (Koh & Roundtree, 2005)
• an algorithm for mining closed association rules (Szathmary et al., 2006)
• an algorithm for mining minimal non-redundant association rules (Kryszkiewicz, 1998)
• the Indirect algorithm – indirect association rules (Tan et al., 2000; Tan et al., 2006)
• the FHSAR algorithm – hiding sensitive association rules (Weng et al., 2008)
• the TopKRules algorithm – top-k association rules (Fournier-Viger, 2012, 📊 slides)
• the ETARM algorithm – top-k association rules (Nguyen et al., 2017) new
• the FTARM algorithm – top-k association rules (Liu et al., 2019) new
• the TopKClassRules algorithm – top-k class association rules (Fournier-Viger, 2012, 📊 slides)
• the TNR algorithm – top-k non-redundant association rules (Fournier-Viger, 2012, 📊 slides)
• the HGB and HGB_All – high-utility association rules (Sahoo et al., 2015) new
• algorithms for mining class association rules
  • the ACAC algorithm (Huang et al., 2011)
  • the ACCF algorithm (Li et al., 2008)
  • the ACN algorithm (Kundu et al., 2008)
  • the ADT algorithm (Wang et al., 2000)
  • the CBA algorithm (Liu et al., 1998)
  • the CBA2 algorithm (Liu et al., 2001)
  • the CMAR algorithm (Li et al., 2001)
  • the L3 algorithm (Baralis et al., 2002)
  • the MAC algorithm (Abdelhamid et al., 2012)

Stream Pattern Mining

These algorithms discover various kinds of patterns in a stream (an infinite sequence of database records).

• the estDec algorithm – recent frequent itemsets in a data stream (Chang & Lee, 2003)
• the estDec+ algorithm – recent frequent itemsets in a data stream (Shin et al., 2014)
• the CloStream algorithm – frequent closed itemsets in a data stream (Yen et al., 2009)
• algorithms for mining the top-k high-utility itemsets from a data stream
  • the FHMDS and FHMDS-Naive algorithms (Dawar et al., 2017)

Clustering

These algorithms automatically find clusters in different kinds of data.

• the original K-Means algorithm (MacQueen, 1967)
• the Bisecting K-Means algorithm (Steinbach et al., 2000)
• the K-Means++ algorithm (Arthur et al., 2007) new
• algorithms for density-based clustering
  • the DBScan algorithm (Ester et al., 1996)
  • the Optics algorithm - extract a cluster ordering of points, which can then be use to generate DBScan style clusters and more (Ankerst et al., 1999)
  • the Density Peak Clustering (DPC) algorithm (Rodriguez et al., 2014)
  • the AEDBScan algorithm (Mistry et al., 2021)
• a hierarchical clustering algorithm
• a tool called Cluster Viewer for visualizing clusters
• a tool called Instance Viewer for visualizing the input of clustering algorithms

Time Series Mining

These algorithms perform various tasks to analyze time series data.

• converting a time series to a sequence of symbols using the SAX representation of time series. Note that if one converts a set of time series with SAX, he will obtain a sequence database, which allows to then apply traditional algorihtms for sequential rule mining and sequential pattern mining on time series  (SAX, 2007)
• calculating the prior moving average (noise removal) of a time series
• calculating the cumulative moving average (noise removal) of a time series
• calculating the central moving average (noise removal) of a time series
• calculating the median smoothing (noise removal) of a time series
• calculating the exponential smoothing (noise removal) of a time series
• calculating the min-max normalization of a time series
• calculating the autocorrelation function of a time series
• calculating the standardization of a time series
• calculating the first and second order differencing of a time series
• calculating the piecewise aggregate approximation (data point reduction) of a time series
• calculating the linear regression (least squares method) of a time series
• splitting a time series into segments of a given length
• splitting a time series into a given number of segments
• clustering time series(group time-series according to their similarities). This can be done by applying the clustering algorithms offered in SPMF (K-Means, Bisecting K-Means, DBScan, OPTICS, Hierarchical clustering) on time series.
  

• a tool called Time Series Viewer for visualizing time series

Classification

• the ID3 algorithm for building decision trees (Quinlan, 1986)
• the KNN (K-Nearest Neighbor) algorithm
• classification based on class association rule mining
  • the ACAC algorithm (Huang et al., 2011)
  • the ACCF algorithm (Li et al., 2008)
  • the ACN algorithm (Kundu et al., 2008)
  • the ADT algorithm (Wang et al., 2000)
  • the CBA algorithm (Liu et al., 1998)
  • the CBA2 algorithm (Liu et al., 2001)
  • the CMAR algorithm (Li et al., 2001)
  • the L3 algorithm (Baralis et al., 2002)
  • the MAC algorithm (Abdelhamid et al., 2012)
• a framework for comparing multiple classifiers using holdout and k-fold cross-validation

Text Mining

• an algorithm for classifying text documents using a Naive Bayes classifier approach (S. Raghu, 2015)
• an algorithm for clustering texts using the tf*idf measure (S. Raghu, 2015)

Dataset Generation Tools

• A tool for generating a synthetic transaction database
• A tool for generating a synthetic sequence database
• A tool for generating a synthetic sequence database with timestamps
• A tool for generating datasets for clustering

Dataset Transformation Tools

• A tool for converting a sequence database to a transaction database
• A tool for converting a transaction database to a sequence database
• A tool for converting a text file to a sequence database (each sentence becomes a sequence)
• A tool for converting a sequence database in various formats (CSV, KOSARAK, BMS, IBM…) to SPMF format
• A tool for converting a transaction database in various formats (CSV…) to SPMF format
• A tool for converting time-series to a sequence database
• A tool to generate utility values for a transaction database
• A tool to add timestamps to a sequence database
• A tool to fix a transaction database having problems (with or without utility/time information)
• A tool for removing utility information from a database having utility information
• A tool to resize a database in SPMF format using a percentage of lines
• A tool to sample records from a dataset (reservoir, seed, etc.)
• A tool to remove duplicated records from a dataset

Dataset Statistics Tools

• A tool for calculating statistics about a transaction database
• A tool for calculating statistics about a transaction database with utility information
• A tool for calculating statistics about a sequence database
• A tool for calculating statistics about a graph database
• A tool for calculating statistics about a product transaction database new
• A tool for calculating statistics about a sequence database with cost and binary utility new
• A tool for calculating statistics about a sequence database with cost and numeric utility new
• A tool for calculating statistics about a sequence database with utility new
• A tool for calculating statistics about a time-extended sequence database new
• A tool for calculating statistics about a transaction database with cost and utility new
• A tool for calculating statistics about a transaction database with utility and period information new
• A tool for calculating statistics about a transaction database with utility and timestamps new
• A tool for calculating statistics about an event sequence new
• A tool for calculating statistics about an interval sequence database new
• A tool for calculating statistics about a multi-dimensional sequence database new
• A tool for calculating statistics about a multi-dimensional sequence database with timestamps new
• A tool for calculating statistics about an uncertain transaction database new
• A tool for calculating statistics about a file with double vectors (instances) for clustering
• A tool for calculating statistics about time series

Dataset Viewer Tools

• A time series viewer to visualize time series
• A cluster viewer to visualize clusters produced by clustering algorithms
• A graph viewer to view files containing graphs or subgraphs (TKG, gSpan, cgSpan)
• A simple tool to view the content of an ARFF file new
• A tool to view the content of an event sequence file new
• A tool to view a sequence database cost binary utility file new
• A tool to view a sequence database cost numeric utility file new
• A tool to view a sequence database file new
• A tool to view a time-extended sequence database new
• A tool to view a multi-dimensional sequence database new
• A tool to view a multi-dimensional time sequence database new
• A tool to view a sequence utility database file new
• A tool to view a cost utility transaction database file new
• A tool to view a transaction database file new
• A tool to view an uncertain transaction database file new
• A tool to view a utility transaction database file new
• A tool to view a utility time transaction database file new
• A tool to view a utility period transaction database file new
• A tool to view a product transaction database file new
• A tool to view a graph database file
• A tool to view a sequence database file with time intervals new
• A tool to view a taxonomy file new

GUI Tools

• The Algorithm Explorer tool to explore the algorithms offered in SPMF
• The Memory Viewer tool to observe the memory usage of algorithms in real-time new
• The Pattern Viewer tool to view patterns found by algorithms and their frequency distributions
• The Workflow Editor tool to create a workflow with several algorithms and run it new
• A tool to run experiments where one or more algorithms are run and a parameter is varied
• The SPMF text editor
• A tool to download an offline copy of the SPMF documentation new
• A tool called Pattern Diff Analyzer to compare two files of patterns to find contrast patterns new
• A tool called Algorithm Graph Viewer to view the similarity between algorithms as a graph new

Other Tools

• A tool to export the list of algorithms to a JSON file new

Data Structures

• Red-black tree
• Itemset-tree
• Binary tree
• KD-tree
• Triangular matrix
• A collection of optimized primitive-type data structures to replace hashmaps, lists, sets, etc.