Complete Closed Time Intervals-Related Patterns Mining

Authors: Omer David Harel, Robert Moskovitch4098-4105

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental For the evaluation of TIRPClo we used several real-world datasets from various domains. The main properties of the datasets are summarized in Table 1 considering five parameters: |E| the number of entities; |STIs| the total number of STIs; |STIs|/|E| the mean number of STIs per entity; MHS the mean horizontal support of a symbol within an entity; and |S| the number of symbol types in the dataset. Datasets extreme conditions appear in bold. ... Experiment 2. Runtime Duration Comparison In this experiment TIRPClo was compared to CCMiner (Chen, Weng, and Hui 2016), which is the only current method designed for closed TIRPs discovery.
Researcher Affiliation Academia Omer David Harel, Robert Moskovitch Software and Information Systems Engineering, Ben Gurion University of the Negev, Be er Sheva, Israel omerdavi@post.bgu.ac.il, robertmo@bgu.ac.il
Pseudocode Yes Algorithm 1: TIRPClo... Algorithm 2: Extend TIRP... Algorithm 3: ESProject
Open Source Code Yes 4. The code of the TIRPClo algorithm as well as the evaluation datasets, which are all publicly available1. 1https://github.com/omerh18/TIRPClo
Open Datasets Yes 4. The code of the TIRPClo algorithm as well as the evaluation datasets, which are all publicly available1. 1https://github.com/omerh18/TIRPClo
Dataset Splits No The paper uses
Hardware Specification Yes All methods were implemented in Visual C# and the experiments were conducted on a dell G5, having 16GB main memory, running Microsoft Windows 10.
Software Dependencies No The paper states that
Experiment Setup Yes using minimum vertical support thresholds of 10%-70% and a fixed maximal gap value of 20. ... the smart-home dataset was used with maximal gap values between 10 40, having the minimum vertical support value fixed on each of the following thresholds: 20%, 30% and 40%.