Web Demo

Several programs inside the IlliMine package have web interfaces for demonstration. The list of available demos is shown to the right. Descriptions are available below.

To run the demos properly, you must enable JavaScript in your browser. If you have any problems running the demos, either post your questions to the mailing list or contact us directly.

Descriptions

More detailed descriptions of the algorithms are available here. Other items such as READMEs and publications are available in the resources section.

Data Cubing

• BUC -- Bottom-up computation of data cube (full or iceberg). Demo includes a parametric data generator.
• Star-Cubing -- Computation of data cube (full or iceberg) through the use of the Star-Tree. Demo includes a parametric data generator.
• MM-Cubing -- Density-based cube computation method for either full or iceberg cubes. Demo includes data generator.
• C-Cubing -- Cube computation using aggregation closedness-checking for efficient calculation of the closed cube. Demo includes the data generator.

Association Mining

• FPGrowth -- Computes the set of frequent itemsets in a dataset with respect to some minimum support threshold. A parametric data generator is included.
• ParFP (Sequential) -- Another implementation of FPGrowth.
• CLOSET+ -- Computes the set of closed frequent itemsets in a database with respect to some minimum support (count) threshold. Data generator is included.
• RPMine -- Computes the set of compressed frequent itemsets in a database with respect to some minimum support (count) threshold. Four test datasets are provided.

Sequential Mining

• PrefixSpan -- Given a set of sequences, computes the set of frequent subsequences with respect to a minimum support threshold. Data generator is included.
• CloSpan -- Given a set of sequences, computes the set of closed frequent subsequences with respect to a minimum support threshold. Data generator is included.
• IncSpan -- Computes the set of frequent subsequences in a sequential database that is incrementally updated. Demo includes the data generator for simulating incremental changes.

Other

• gSpan -- Computes the set of frequent subgraphs in a graph dataset. Demo includes 2 different datasets.
• CPAR -- Classification based on predictive association rules. Demo allows CPAR to run on the UCI datasets.