KDD procedures and modules

There are procedures for input data transformations, several data mining procedures based on the GUHA principle and machine learning procedure KEX. Some of procedures have additional modules.

All procedures and modules need subsystem Elementary. Both data mining procedures and machine learning procedure KEX have important common features.

4ft-Miner

Data mining procedure 4ft-Miner mines for various types of association rules. Rules based on statistical hypothesis test and conditional association rules are also included. There is a demonstration of the main features of the procedure 4ft-Miner.

4ft-IL

4ft-IL is an additional module of the 4ft-Miner procedure. It allows analysis of importance particular literals of found association rules.

4ft-UF

4ft-UF is an additional module of the 4ft-Miner procedure. It allows sorting of found association rules according to individual preferences of interestingness.

4ft-AR2NL

4ft-AR2NL is an additional module of the 4ft-Miner procedure. It automatically converts some of association rules into natural language form.

SD4ft-Miner

Data mining procedure SD4ft-Miner mines for couples of sets that differ significantly in respect to some association rule.

KL-Miner

Data mining procedure KL-Miner mines for patterns based on evaluation of two-dimensional contingency tables.

KL-Collaps

KL-COLLAPS is an additional module of the KL-Miner procedure. It allows deeper analysis of found patterns concerning two – dimensional contingency table.

CF-Miner

Data mining procedure CF-Miner mines for conditions under which frequencies of categories of an attribute satisfy given requirements.

SDKL-Miner

Data mining procedure SDKL-Miner mines for couples of sets that differ significantly in respect to a pair of categorial attributes under some condition.

SDCF-Miner

Data mining procedure SDCF-Miner mines for couples of sets that differ significantly in respect to conditional frequencies of categories of an attribute under some condition.

Action4ft-Miner

Data mining procedure Action4ft-Miner mines for action rules.

ETree-Miner

Data mining procedure ETree-Miner mines for exploration-tree structures.

KEX

Machine learning procedure KEX (Knowledge EXplorer) performs symbolic empirical multiple concept learning from examples, where the induced concept description is represented as a set of weighted decision rules.

Data Preprocessing

The Data Preprocessing subsystem consists of empty metabase, module Admin and of data exploration and transformation procedure DataSource.

TimeTransf

Data transformation procedure TimeTransf computes various characteristics of time series. The resulting characteristics can be used as input of analytical procedures.

LM TaskPooler

Batch processing of tasks in the background

LM GridPooler

Batch processing of tasks on the grid

LM Reverse-Miner

The Reverse-Miner module generates data that could be later analysed in the usual way by the LISp-Miner system (or any other system for KDD).

LM Knowledge

LispMiner Modules concerning KnowledgeBase and LAQs

Sewebar

Import and export of the PMML documents into and out of the LM Metabase. Used primary for data interchange with the SEWEBAR project but allows for export of other formats as well.

Miscellaneous

Other files for download