Reproducibility artifacts for "Scatter or Intersect? An Experimental Analysis of Counting Strategies for Low-Support Frequent Itemset Mining"
收藏资源简介:
Complete reproducibility package for the fastapriori (HybridFIM) study of counting strategies for low-support frequent itemset mining. It contains library source, comparator sources, the measurement harness, saved results of all campaigns, notebooks, and the input data. Notebook 21 regenerates the study's tables, figures and quoted numbers from the saved results without any mining. The archive comes in four zip parts. Download all four and extract them into one destination folder; they merge into fastapriori-artifacts-0.2.1/. On Linux or macOS run: unzip 'fastapriori-artifacts-0.2.1-part*.zip'. Windows "Extract All" suggests a separate folder for each zip, so choose the same destination each time. Then start with README.md. Instacart is not included: download it from Kaggle (https://www.kaggle.com/datasets/psparks/instacart-market-basket-analysis) and follow data/SOURCES.md. The other seven real datasets and all synthetic datasets are included. Code (minus the data) at https://github.com/swetmr/fastapriori/tree/main/artifacts Licences: the author's code is MIT and the author's results and documentation are CC BY 4.0. Third-party datasets and comparator sources keep their own terms, listed in data/SOURCES.md and in the tools folder. The licences selected on this record do not cover them.



