RapidMiner

from Wikipedia, the free encyclopedia
RapidMiner

A screenshot from RapidMiner (YALE)
A screenshot from RapidMiner (YALE).
Basic data

developer rapidminer.com
Current  version 9.6
(March 2, 2020)
operating system Platform independent
programming language Java
category Artificial intelligence
License proprietary and AGPL with reduced functionality
German speaking Yes
Project page

RapidMiner , previously called YALE ( Yet Another Learning Environment ), is an environment for machine learning and data mining . Experiments can be generated from a large number of operators that can be nested as required. The experimental setup is described by XML and developed using a graphical user interface. RapidMiner covers research as well as industrial and economic applications.

YALE was developed since 2001 by the chair for artificial intelligence at the Technical University of Dortmund . YALE has been hosted by SourceForge since 2004 . In May 2007, YALE was renamed RapidMiner ; Version 4.0 was released in July 2007 and Version 4.3 in November 2008. RapidMiner is available as a proprietary version, as well as the respective previous version as open source software under the AGPL on Sourceforge. In the meantime, however, the free version has been limited to only 10,000 data records and a logical processor, which means that use is very limited and, comparable to the earlier versions - which could still use all existing cores - is very time-consuming. In addition, a lot of functionality has been moved to "extensions" that are no longer part of the open source version.

RapidMiner contains more than 500 operators for all tasks of knowledge discovery in databases , i. H. Operators for input and output, data preprocessing, machine learning , data mining , text mining , web mining , automatic sentiment analysis from Internet discussion forums (sentiment analysis, opinion mining), time series analysis and forecasting. In addition, more than 20 methods are available to visualize high-dimensional data and models. RapidMiner was written in Java and can therefore be used on all common operating systems. All learning processes and attribute evaluators from WEKA have also been integrated.

Companies

At the end of 2013, RapidMiner received funding of $ 5 million from Earlybird and Open Ocean Capital (the VC of the MySQL developers). The capital was collected and used for the US market strategy. RapidMiner has a company headquarters in Boston . In 2013, the number of employees rose from 25 to 70. Sales growth was 300 percent year-on-year. In 2016 the company employed 100 people, 30 of them in Dortmund.

properties

Some features of RapidMiner are:

  • Knowledge discovery processes are modeled as operator trees
  • Internal XML representation ensures a standard for the exchange of data mining experiments
  • Script language allows automated large-scale data mining
  • Multi-layered data view concept ensures efficient and transparent data storage
  • Graphical user interface and command line tool. A Java API allows RapidMiner to be used from your own Java programs
  • Plugin and extension mechanism, some plugins already exist, e.g. B. for cluster analysis
  • Large number of high-dimensional visualizations for data and models
  • Applications include text mining, multimedia mining, feature engineering, data stream mining and changing concepts learning, ensemble method development and distributed data mining.

See also

Web links

Individual evidence

  1. ↑ The core of RapidMiner is open source . RapidMiner. Retrieved July 18, 2014.
  2. ↑ http://www.gruenderszene.de/allgemein/rapidminer-earlybird
  3. ↑ http://www.gruenderszene.de/allgemein/ingo-mierswa-rapidminer-startup-helden
  4. ↑ http://wirtschaftsfoerderung-dortmund.de/service/presse/pressedetailsseite_18304.html