A Survey of High Utility Sequential Pattern Mining

The problem of mining high utility sequences aims at discovering subsequences having a high utility (importance) in a quantitative sequential database. This problem is a natural generalization of several other pattern mining problems such as discovering frequent itemsets in transaction databases, frequent sequences in sequential databases, and high utility itemsets in quantitative transaction databases. To extract high utility sequences from a quantitative sequential database, the sequential ordering between items and their utility (in terms of criteria such as purchase quantities and unit profits) are considered. High utility sequence mining has been applied in numerous applications. It is much more challenging than the aforementioned problems due to the combinatorial explosion of the search space when considering sequences, and because the utility measure of sequences does not satisfy the downward-closure property used in pattern mining to reduce the search space. This chapter introduces the problem of high utility sequence mining, the state-of-art algorithms, applications, present related problems and research opportunities. A key contribution of the chapter is to also provide a theoretical framework for comparing upper-bounds used by high utility sequence mining algorithms. In particular, an interesting result is that an upper-bound used by the popular USpan algorithm is not an upper-bound. The consequence is that USpan is an incomplete algorithm, and potentially other algorithms extending USpan.

[1]  Johannes Gehrke,et al.  Sequential PAttern mining using a bitmap representation , 2002, KDD.

[2]  Philip S. Yu,et al.  Mining high utility episodes in complex event sequences , 2013, KDD.

[3]  Vincent S. Tseng,et al.  Mining High Utility Sequential Patterns from Evolving Data Streams , 2015 .

[4]  Aijun An,et al.  Top-k utility-based gene regulation sequential pattern discovery , 2016, 2016 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[5]  Hoai Bac Le,et al.  Efficient algorithms for simultaneously mining concise representations of sequential patterns based on extended pruning conditions , 2018, Eng. Appl. Artif. Intell..

[6]  J. Roberts,et al.  HAWAIIAN SKIRT controls size and floral organ number by modulating CUC1 and CUC2 expression , 2017, PloS one.

[7]  Mohammed J. Zaki,et al.  SPADE: An Efficient Algorithm for Mining Frequent Sequences , 2004, Machine Learning.

[8]  Jianliang Xu,et al.  Mining High Utility Sequential Patterns with Negative Item Values , 2017, Int. J. Pattern Recognit. Artif. Intell..

[9]  Hoai Bac Le,et al.  A Novel Approach for Hiding High Utility Sequential Patterns , 2015, SoICT.

[10]  Uma Dave,et al.  Efficient Mining of High Utility Sequential Pattern from Incremental Sequential Dataset , 2015 .

[11]  Vincent S. Tseng,et al.  Efficient Mining of High-Utility Sequential Rules , 2015, MLDM.

[12]  Jerry Chun-Wei Lin,et al.  Mining of high utility-probability sequential patterns from uncertain databases , 2017, PloS one.

[13]  Byeong-Soo Jeong,et al.  A Novel Approach for Mining High‐Utility Sequential Patterns in Sequence Databases , 2010 .

[14]  Yi-Cheng Chen,et al.  On efficiently mining high utility sequential patterns , 2016, Knowledge and Information Systems.

[15]  Van-Nam Huynh,et al.  Mining Periodic High Utility Sequential Patterns , 2017, ACIIDS.

[16]  Philippe Fournier-Viger,et al.  High-Utility Sequential Pattern Mining with Multiple Minimum Utility Thresholds , 2017, APWeb/WAIM.

[17]  Tzung-Pei Hong,et al.  Applying the maximum utility measure in high utility sequential pattern mining , 2014, Expert Syst. Appl..

[18]  Pinar Senkul,et al.  CRoM and HuspExt: Improving Efficiency of High Utility Sequential Pattern Extraction , 2015, IEEE Transactions on Knowledge and Data Engineering.

[19]  Philippe Fournier-Viger,et al.  FCloSM, FGenSM: two efficient algorithms for mining frequent closed and generator sequences using the local pruning strategy , 2017, Knowledge and Information Systems.

[20]  Longbing Cao,et al.  Efficiently Mining Top-K High Utility Sequential Patterns , 2013, 2013 IEEE 13th International Conference on Data Mining.

[21]  Byeong-Soo Jeong,et al.  Mining High Utility Web Access Sequences in Dynamic Web Log Data , 2010, 2010 11th ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing.

[22]  Sylvie Norre,et al.  TWINCLE : A Constrained Sequential Rule Mining Algorithm for Event Logs , 2017, KES.

[23]  Longbing Cao,et al.  USpan: an efficient algorithm for mining high utility sequential patterns , 2012, KDD.

[24]  Manuel Campos,et al.  Fast Vertical Mining of Sequential Patterns Using Co-occurrence Information , 2014, PAKDD.

[25]  Ying Liu,et al.  A Two-Phase Algorithm for Fast Discovery of High Utility Itemsets , 2005, PAKDD.

[26]  Vincent S. Tseng,et al.  A One-Phase Method for Mining High Utility Mobile Sequential Patterns in Mobile Commerce Environments , 2012, IEA/AIE.

[27]  Yonggang Hu,et al.  Distributed and parallel high utility sequential pattern mining , 2016, 2016 IEEE International Conference on Big Data (Big Data).

[28]  Roque Marín,et al.  ClaSP: An Efficient Algorithm for Mining Frequent Closed Sequences , 2013, PAKDD.

[29]  Philip S. Yu,et al.  Mining interesting user behavior patterns in mobile commerce environments , 2012, Applied Intelligence.

[30]  Philip S. Yu,et al.  Mining High Utility Mobile Sequential Patterns in Mobile Commerce Environments , 2011, DASFAA.