8
Data Mining and Predictive Analytics Toolkit December 2013, Jakub Miarka, University of Leeds

Data Mining and Predictive Analytics Toolkit

Embed Size (px)

DESCRIPTION

Data Mining and Predictive Analytics Toolkit. December 2013, Jakub Miarka, University of Leeds. RapidMiner. Formerly called YALE (Yet Another Language Environment) Environment for machine learning, data and text mining, predictive and business analytics - PowerPoint PPT Presentation

Citation preview

PowerPoint Presentation

Data Mining and Predictive Analytics ToolkitDecember 2013, Jakub Miarka, University of Leeds

1RapidMinerFormerly called YALE (Yet Another Language Environment)Environment for machine learning, data and text mining, predictive and business analyticsStarted in 2001 at the Artificial Intelligence Unit of the Dortmund University of Technology, Germany2006 Rapid-I foundedAGPL open source license until November 2013Profitable company, growing organically3 millions downloads / 200,000 usersOne of the leaders in the predictive analytics

UsageGUI for building data mining/analytics workflowsHighly scalable predictive analytics applicationLearning schemes and attribute evaluators from WEKAIntegrates with popular enterprise data sources (60+, incl. SAP) Supports both structured and unstructured dataTypically used for:customer segmentationloyalty and retention analysiscredit ratingsasset maintenanceresource planning

A pie chart showing aggregated informationMultiple results displayed simultaneously

BenefitsNo programming skills needed and easy to use (GUI, drag & drop)1000+ analytical methods120+ models incl. decision trees and dozens of visualisations availablePowerful and scalableFlexible, scriptable, supports plugins and extensionsProvides a GUI to design an analytical pipeline (the "operator tree") which defines the analytical processes the user wishes to apply to the dataOther applications can use the engine through API

PopularitySuitable for individuals and large enterprises as wellSome of the customers:PayPalPepsiCoeBayVolkswagenLufthansa and many more

November 2013 and future$5 millions investmentRebranded from Rapid-I to RapidMinerCore stays open source but new commercial packages introducedWhen a new version is published, previous ones become freeIn future, increased focus on Big Data and self-service-style interface for less technical and more business-focused usersA vision to become the industry standard for predictive analytics

Referenceshttp://rapidminer.com/products/rapidminer-studio/http://sourceforge.net/p/rapidminer/wiki/Home/http://en.wikipedia.org/wiki/RapidMinerhttp://techcrunch.com/2013/11/04/german-predictive-analytics-startup-rapid-i-rebrands-as-rapidminer-takes-5m-from-open-ocean-earlybird-to-tackle-the-u-s-market/http://www.zdnet.com/rapid-i-gets-funded-re-brands-as-rapidminer-7000022757/