Constraint-Based Sequential Pattern Mining with Decision Diagrams
Authors: Amin Hosseininasab, Willem-Jan van Hoeve, Andre A. Cire1495-1502
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To evaluate the applicability of our approach, we develop an MDD-based prefix-projection algorithm and compare its performance against a typical generate-and-check variant, as well as a state-of-the-art constraint-based sequential pattern mining algorithm. Results show that our approach is competitive with or superior to these other methods in terms of scalability and efficiency. |
| Researcher Affiliation | Academia | Amin Hosseininasab,1 Willem-Jan van Hoeve,1 Andre A. Cire2 1Tepper School of Business, Carnegie Mellon University, USA 2Dept. of Management, University of Toronto Scarborough, Canada |
| Pseudocode | No | The paper describes the steps of the MDD prefix-projection (MPP) algorithm in text, but it does not provide formal pseudocode or an algorithm block. |
| Open Source Code | Yes | The MPP code is available and open source.4 [Footnote 4: https://github.com/aminhn/MPP] |
| Open Datasets | Yes | For our numerical tests, we use real-life click-stream benchmark databases2, listed in Table 3. [Footnote 2: http://www.philippe-fournierviger.com/spmf/index.php?link=datasets.php] |
| Dataset Splits | No | The paper refers to using "minimum support (Min supp) as a percentage of the total number of sequences" but does not specify a distinct training, validation, or test split for the datasets. |
| Hardware Specification | Yes | All experiments are executed on the same PC with an Intel Xeon 2.33 GHz processor, 24GB of memory, using Ubuntu 12.04.5 as operating system. We limit all tests to use one core of the CPU. |
| Software Dependencies | No | All algorithms are coded in C++, with the exception of PPICt which is coded in Scala. No specific versions for C++ compilers, Scala, or any libraries/frameworks are mentioned. |
| Experiment Setup | Yes | The experiment uses three scenarios with constraints on one, two, and three attributes, respectively: time: 30 Cgap(time) 900, 900 Cspn(time) 3600, price: 30 Cavg(price) 70, 40 Cmed(price) 60, quality: 40 Cavg(quality) 60, 30 Cmed(quality) 70. |