Dealing with Background Knowledge in the SEWEBAR Project

EverMiner - towards Fully Automated KDD Process

“…Although the above-mentioned Knowledgebase was implement first in the LM DataSource and the LM KnowledgeSource modules of the LISp-Miner system (see [Rauch & Šimůnek, 2008] and [LISp-Miner]) its future development was moved now into the SEWEBAR project.…”

“…Not only task parameters but also types of patterns the automated process is looking for could be influenced by previous results and changed during iterations. This text specifies further and in more details the thoughts published in previous works of [Rauch & Šimůnek, 2005a], [Rauch & Šimůnek, 2007], [Rauch & Šimůnek, 2009a] and mainly in the paper where a formal architecture of the automated data-mining process was firstly proposed. The text is organized as follows.…”

“…Distributed solving of data mining tasks using the computer grid was implemented (see [Šimůnek & Tammisto, 2010]). The first version of the Knowledgebase for storing domain knowledge has been built within the LISpMiner LM DataSource module (see [Rauch & Šimůnek, 2008]) and is now refined further in cooperation with the SEWEBAR project (see e.g. [Kliegr et.…”

SEWEBAR-CMS: semantic analytical report authoring for data mining results

“…Together with accompanying XML formats and workflow, SEWEBAR-CMS is a part of the SEWEBAR framework [38] for analytical report authoring.…”

“…-Meta-attributes: The basic building block of BKEF is a Meta-attribute [38]: an abstraction representing a property appearing in datasets from the given domain. The definition of meta-attributes can be used e.g.…”